How about if you injure yourself playing sports? Or if your job involves exposure to hazards (hey, you didn't have to take that job!)?

Isn't workplace comp supposed to cover (both of) those occurrences?

If someone breaks their legs playing sports... shouldn't the sports organizations cover it, instead of everyone in the USA subsidizing that particular entertainment organization? Otherwise, what incentive would cause them to make the sport safer or introduce better sport equipment?
Likewise, if someone works at a chemical plant, shouldn't the chemical plant be responsible for covering health costs related to it, or else why would they bother making things safer?

Breaking your leg on your front doorstep, or stepping off a street curb, is different. That's an issue for general healthcare, unless something is directly wrong with the curb, in-which case you have to figure out who is responsible for that (homeowner insurance, or what), and how much they should bear the burden.

'course consumers shouldn't have to deal with that ("Who's paying for my injury? Who's at fault? (if anyone)"), so healthcare should automatically cover it (even the workplace issues), but the health-insurance provider reimbursed by whoever is actually liable.

The reason why it's important that we identify who is "actually liable" (or whether it's just a general accident like tripping and falling), is to make sure someone is incentived to fix the problem if there actually is a problem.

This reaches into the realms of absurdity, but let's reach a little further to illustrate where this train of thought leads us:
A) If you drink lots of soda and eat hamburgers, you get fat.
B) Fat people are very unhealthy and generally have a lower life expectancy.
C) Therefore, fat people will cause a burden on the health care system because they need extra health care.

[...]

Logically, A is therefore bad because of C; But morally, how much right do we have to dictate the lifestyle choices and habits of others if and only if C is marginally* harmful? (*accounting for people who choose A to be a habit of serial killing).

We should thus recognize that all of this hinges upon our definition of "marginally harmful", and that is somewhat of a subjective measurement. So, maybe the level of restriction on A should be proportionate to the harm done by C, as agreed to by a diverse cross section of society which will bear the costs of C?
ie, drinking and driving is bad, so don't drink AND drive, but you can drive or drink, but not both.

If obesity is a burden on society in general (through shared health costs), then that behavior ought to be taxed higher. Instead of: "If your doctor says you are obese, you have to pay 5% extra health insurance", we just ought to tax fast food companies that specifically serve unhealthy food, and soda companies with unhealthy drinks, and things like that. The cost increases will be naturally passed on to the consumers eating the food, reducing the amount they eat (or else increasing the amount they pay), or else the fast food companies will be incentived to make the food healthier, because real hamburgers aren't actually unhealthy! (it's just bread and meat and vegetables) Nor are french fries unhealthy (just potatoes). It's how it's grown, processed, preserved, and cooked that makes it unhealthy.

I haven't actually studied these issues, so I might be way offbase, but my first-approximation approach would be something like:

A) The taxes for "burdensome on society" items should actually go to alleviating that burden. e.g. tobacco taxes must be go a state fund to compensate health insurance providers on specifically tobacco-related health problems, not general state budgets, or the taxes will just be an easy way for governors and politicians to make up budget shortfalls from their financial incompetence, and will just get raised higher and higher to cover increasingly large budget shortfalls.

B) The states must work together to set taxes reasonably consistently, or the product will be smuggled between states (like how cigarettes get smuggled into New York from nearby states, because the New York taxes are much higher (by several bucks a pack) than surrounding states).

C) The taxes should be automatically adjusted based on the actual burden of the previous year. If total tobacco-related government health-cost reimbursements in 2020 were 7% higher than total tobacco-related taxes in 2020, taxes for 2021 should automatically be adjusted to match. Likewises if total tobacco-related taxes were higher than total tobacco-related health costs, the taxes for 2021 should be automatically decreased. Otherwise, tobacco-smokers will feel punished for smoking, instead of merely paying their fair share (via tobacco-product taxes). If it turns out tobacco-related health issues actually aren't that big of an issue, it'll automatically adjust over a year or two.

The biggest difference between obesity and drinking, verses smoking, is smoking is also harmful to people around the smoker.
While my solution above will automatically cover the health costs of second-hand smoke, it doesn't cover the misery of it. If someone never has had to go to the hospital for any reason, and they really watch their health, but they have to have surgery (or whatever) for secondhand (or thirdhand) smoke, and merely having your health costs paid for does nothing to compensate you for your now permanently decreased health, lost days of work/life/whatever, and the misery of the surgery itself; and that misery is directly caused from other people's choices. The same, on a more extreme level, is drunk-driving accidentally killing someone (with the 'choice' being driving while drunk).

I agree that there are serious health effects from smoking cigarettes, cigars, hookahs, etc., but those aren't the direct reason for the bans.
[...]
I also find tobacco smoke unpleasant and believe that health dangers of secondhand smoke are real.

Those are the real reasons why it's banned in public places. Smoking does not just hurt yourself (that's mostly your problem, who cares) but also the people around you.

Alcohol is banned from traffic for exactly that reason, too. Nobody cares if you drink and drive and end up dead on the next tree1. On the contrary, that's one more organ donor, and we need organs, so you're actually doing something for society. But the problem is, you risk running over a pedestrian or hitting another car (or think a bus with 20 people) and kill non-drinkers.

What if... where to draw the line?

The line is drawn by insurance companies, or in the case of public health insurance, by the government. It needs not always be perfectly logical, but there is some logic behind it. In the case of insurance companies, they simply count the number of incidences during the last 20 years, round up to the next ten percent, and calculate your fees from that. Then they multiply the result with 1.5.

There is an important difference to note. Drinking can kill you, but moderate drinking will usually not have that effect. Eating sugar can cause diabetes, but needs not, and in moderate doses will not. On the other hand, the effects of smoking are, although denied by the tobacco industry for obvious reasons, undeniably always desastrous. There is no threshold, no harmless dose, and there is no beneficial dose either.

Given enough time, nature makes the best out of most toxic/harmful things, or at the very least it finds a way to mitigate the adverse effects (Why? well simply because the ones who can't cope die early!). Sunlight is extremely harmful, as is oxygen. We have dedicated systems that repair the DNA damage done by UV radiation and oxygen, and our skins have a built-in adaptive sunshield. Oxygen is the main component in our energy supply (and a waste product generated by plants, to whom the dangerous UV radiation is the main source of energy). Alcohol is quite harmful, but we have dedicated enzymes that not only disable the toxic substance, but turn it into a form of energy that our cells can use.

Now... nicotine... is a strong neurotoxin, so strong you can consider it chemical warfare. It's nature's form of sarin (well, not really... sarin inhibits the cholinesterase while nicotin directly binds to the cholin receptor -- but in essence it's the same thing). Nicotine serves one purpose for the plant: animals who eat the plant perish. But before you go into discussing at what dose the effects of nicotine may or may not become unhealthy, one should note that the other ingredients of smoke are even worse than the plant poison. Ultra small particule matter that cannot be digested by phagocytes, tar, and non-neglegible amounts of toxic and cancerogeous gases. Including an amount of CO which is so high that you can measure it in blood samples (and thanks to its high affinity, you can still find it the next day, so only smoking every other day is already enough to guarantee a permanent sub-par oxygen supply in organs).

So far, nature's best defense against smoke is that it causes you to caugh and flee from it (at least animals do that, humans sometimes seem to evolve backwards).

One of the biggest problems is that you cannot define a threshold for cancerogenous chemicals. Which is why for example that stupid "Oh, but you have to drink 1000 liters per day" quote which came up with the glyphosate beer story. The point isn't that glyphosate is kinda toxic (it is, too). The point is that it's actively damaging DNA (that is what it's made for, being a weed killer), and thus being a cancerogen and probably a teratogen, too. Sure, you can, and for economic reasons this is done, draw a threshold which is above what you've measured and yell: "Harmless, harmless, harmless!" all day. But that isn't what reality looks like. There is no "harmless" level.

About sugar and obesity, by the way, our Greens are trying to get a sugar tax going. Which is, of course, like all Green ideas total bollocks because the average person has no way of escaping sugar anyway, and it's in virtually everything you buy. If you are being honest (counting starch as sugar) that would be a tax on everything you can buy in the supermarket (except carrots and tomatoes). Thanks to the USA, notably the Nixon administration, there are huge amounts of corn starch in virtually every food. There's particularly high amounts of corn starch in "low fat" and "fit" and "healthy" products, too. Think Philadelphia.

There is of course a huge economic interest in that, too. A lobby ten or twenty times the size of the tobacco lobby (and that's not counting in the pharma lobby which has an obvious interest in people becoming diabetic). So... nothing will come out of that. Nothing useful or sensible, anyway. It will only make everything more expensive. Seeing how the Greens are ultra-left, that's an awesome thing to do, since it's only the poor who will feel food getting more expensive.

1Unless it's a protected tree, or one of a partiuclarly precious kind.

It's really a shame that most people already take for granted that a smoking is bad. Propaganda machine has apparently succeeded in his plan. Smoking ban has a completely different reason. Do your research before commenting and participation in the implementation of someone else's agenda. Tobacco is not bad except additives that are added with cigars. Tobacco is a very useful and healthy plant.