Add feature for animation event animation accessors

[ddbeck]

Fixes #4107

[Elchi3]

Not really sure if these two add-ons are a feature on its own. If they are, I think we should try harder to mention what they offer.

[Elchi3]

Suggested change

	description: The `animation` property of an `AnimationEvent` or `TransitionEvent` object is the `Animation` object that triggered the event.
	description: The `animation` property of an `AnimationEvent` or `TransitionEvent` object represents which Animation (and thus which element) fired a certain animation event.

[ddbeck]

I like your approach and adapted it into c4deaf6.

[ddbeck]

Not really sure if these two add-ons are a feature on its own. If they are, I think we should try harder to mention what they offer.

So I found myself with three plausible paths here:

1. No feature. Add the keys to animation-css and pretend this PR (and the inciting issue) never happened.
2. Mint the feature now, as I've done. Then merge it into animation-css at some point in the future, when have the machinery and this feature passes whatever qualifications we ultimately set for feature merging.
3. Defer this decision entirely. Wait until we get feature merging, then reconsider what to do with these keys then.

I initially wanted to lump it into animation-css. The trouble is that almost everything there is substantially older (e.g., Baseline ~2020 or earlier). I figured that, if we had created this feature in June when it was requested, there would have been no way of knowing that it would be Baseline today and it would have been a more straightforward candidate for a standalone feature (i.e., it would have been more difficult to make the case that we should not mint the feature at all).

If we had a feature merging apparatus already in place, I'd advocate for a fast merge, where a minor extension to an existing feature hits Baseline high means we immediately fold it into its antecedent feature. We could still do that, once we get feature merging in.

Or we could say, yes, it would have been the correct thing to do three months ago, but since we didn't do the correct thing now, we should pretend it never happened in the first place and silently fold it in. That feels a little bit bad.

[Elchi3]

Thanks for elaborating! I think I now agree that mining a feature is the better path forward. Should we leave a comment in the yml that says "Merge into animation-css when this feature becomes baseline high" ?