Also want to highlight that I think I'll need to add additional tests to ensure that style edition is set correctly in different scenarios before this gets merged. Mostly leaving this as a note for myself so I don't forget to add those tests

Whew, this one looks like it'll be fun to review. I assume the changes proposed here are atomic, and that there's no way to decompose it in any form into two (or more) smaller PRs?

If it helps, I think we could break this up a little. All the proc macro code could be one PR, and then there might be some other changes that could be broken up into smaller PRs, but once we start messing with the configuration a lot of it needs to come together.

Let me know what you think would be best. I'm happy to make any changes.

If it helps, I think we could break this up a little. All the proc macro code could be one PR, and then there might be some other changes that could be broken up into smaller PRs, but once we start messing with the configuration a lot of it needs to come together.

Let me know what you think would be best. I'm happy to make any changes.

I think the typical adage about a higher quantity of smaller PRs being preferable to a lower quantity of larger PRs holds here as well. I trust and defer to your judgement around determining the best way to slice and dice the changes

Also, I really want to underscore my appreciation for your efforts on this front. It's obviously no small task, but the work is of paramount importance and I'm grateful you've already made this much progress.

I cannot over stress how significant and beneficial the style edition mechanism is going to be in addressing some incredibly long standing and thorny issues (I will also give yet another shout out to @joshtriplett for coming up with the style edition solution 🙏)

work for this is being done in other PRs