Add ability for maintainers to provide reason for yanking a release

[daneah]

Resolves #7856
Fixes #7886

[daneah]

I'm trying to think about where to add tests...the process of adding a reason and making sure it shows up feels like a functional test, of which there are rather few. This new feature didn't introduce new methods/functions, so maybe I'm also overthinking it. Coverage remains at 100%, but I wouldn't mind exercising the new feature more explicitly.

[daneah]

@di Let me know if you have any guidance on the above notes! Is it okay if I force push a rebase of this branch, or do y'all prefer a fresh PR?

[di]

@daneah This is on my list to review today. A force push is totally fine!

[di]

This looks pretty good! A few thoughts:

• We probably want to unset yanked_reason when the release is unyanked.
• We should surface yanked_reason in the JSON API as well (and add it to the corresponding docs)
• I think the yanked_reason in the JSON API should probably be None when the release is not yanked
• We probably want to display the yanked_reason in the "Yanked Releases" table as well
• We should include yanked_reason in the additional field when we call record_event during yanking

For tests, I would expect to see updates for yanked_reason anywhere where we're doing assert release.yanked or assert not release.yanked to ensure this field is actually getting set in the view. Same for the JSON API tests.

[daneah]

@di thanks for the review! I can definitely address some of those points readily, and I'm sure I'll have a couple of follow up questions 😄 will report back.

[di]

LGTM! I'll give @ewdurbin or @dstufft a chance to review if they want to.

[di]

Thanks @daneah!

[pradyunsg]

Wohoo! Thanks @daneah and @di! ^>^