Commenting on the answer
People vote with their prejudices. There is a point of view where it is unfair that renewable subsidies are counted but the implicit environmental costs are not. People with that point of view applauded the International Monetary Fund for making that assertion because it supports their world view. From their point of view, a lack of externally imposed costs is the equivalent of a subsidy in the same way that imputed rent is income.
That said, the answer that I believe you mean makes that clear in the details. It says, with bolding:
plus a tax to reflect environmental damage
You are of course welcome to disagree with that interpretation. But telling the poster (which is what a comment does) is unlikely to have much effect. The poster knows that this isn't the common definition. But people of that viewpoint don't care. To them, that should be the definition to be fair. The answer explains why at length.
Telling such people that their view is wrong won't change their mind. At best it will express a separate opinion. A better way is to write a separate, cited answer, that expresses the view that "subsidy" is the wrong word.
Of course, if you do that, you're stuck with the fact that on this site, there are more people who favor the environmentalist view that it is unfair not to include externalities as imputed subsidies. So such an answer is likely to receive some downvotes and not so many upvotes.
Evaluating the answer
This meta answer says:
The answer (after a few revisions) went further and showed that this definition was commonly used by a set of experts who were likely to be primary audience of the article.
That is of course absurd. The question wasn't about the original article. The question was about a series of claims that appeared outside environmental science journals. One of them appears in The Guardian, which is not a scientific publication in any way shape or form. So the question was actually whether the statement made sense under the definition used by readers of The Guardian headlines, not in the environmental science literature.
The question also quotes Forbes as saying that the definition is skewed. This too is unsurprising as their prejudices reject the concept of environmental costs. But then the question becomes if it is reasonable to say that the definition is skewed. And of course it is.
That's not opinion-based. In terms of the definition used by Forbes, which is internally consistent, a subsidy is when the government sends a check to someone for that purpose or when the government allows a tax deduction for that purpose that they wouldn't allow for other purposes. By that definition, subsidies for fossil fuels are far smaller.
Note that renewables advocates don't include such subsidies when talking about renewables. They don't include tax deductions taken searching for gallium (used in solar panels) as renewables subsidies. They don't include the environmental damage from dams as renewables subsidies. When they talk about renewables subsidies, they only include direct subsidies and tax breaks. It's only when talking about fossil fuels that they start using a broader definition.
That's not opinion-based. It's just fact. But it's hard to cite, because it goes against the point that environmental scientists are trying to make. It would be difficult to establish, because it would require showing that those costs were not included in the other costs. And it would be hard to publish, because it's not particularly interesting to people who do that kind of work. But it's still hypocrisy.
Beyond all this, it's worth noting that The Guardian uses the term subsidy in two different ways. It claims that subsidies are "largely due to polluters not paying the costs imposed on governments by the burning of coal, oil and gas" and "many of the energy subsidies highlighted by the IMF go toward finding new reserves of oil, gas and coal". On the one side, they claim costs to governments are the bulk of the subsidies and on the other hand they claim that finding new reserves are. That's using subsidies in two different ways in the same article.
That's just hypocrisy. It's not opinion-based to say so.
What is opinion-based is the implicit assumption that all this is justified because most people don't understand things well enough. So the truth has to be distorted so that other people feel as concerned about this as the activists do.
It is reasonable for the IMF to use an internal definition of subsidy. And it is reasonable and not opinion-based to point that out. But that doesn't answer the actual question, which is if the British Medical Journal, The Guardian, and Forbes are also basing their assertions in fact.
Now, I can't look for hypocritical casting in Forbes, because I refuse to allow them to run Javascript and their site doesn't work without it. But their definition of subsidy is consistent with how business people use the word.
This BMJ article is pitched not to economists (as expected from an IMF publication) but to doctors. I'm not sure that the typical doctor defines subsidy as the absence of a Pigovian tax. That argument does not appear in the portion of the article available to non-subscribers.
So in terms of this particular question, whether the IMF definition is internally consistent is beside the point. The real question is if the definitions used by the BMJ, Forbes, and The Guardian are internally consistent. And whether they can be reconciled.
If that's opinion-based, then it is a characteristic of the question itself.