Fix the definition of sigevent on FreeBSD and Linux

[asomers]

It was originally defined back before rust could represent C unions. So instead of defining the union field correctly, it simply defined that union's most useful field. Define it correctly now.

Remove traits that can't be safely implemented on a union: PartialEq, Eq, and Hash. Define Debug, but exclude the union field. Leave Clone in place.

This PR is a rebase of #2813 that eschews all of the backwards-compatibility parts.

[rustbot]

r? @JohnTitor

rustbot has assigned @JohnTitor.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

[JohnTitor]

Failed on musl: https://github.com/rust-lang/libc/actions/runs/8846424497/job/24292411184?pr=3630

[asomers]

@JohnTitor I fixed the build failure on musl and rebased on main

[asomers]

The Sparc failure is not due to this PR.

[SteveLauC]

Considering the main branch is for version 1.0, which allows breaking changes, I think we can move this forward.

[bors]

☔ The latest upstream changes (presumably #3962) made this pull request unmergeable. Please resolve the merge conflicts.

[tgross35]

@asomers mind rebasing this when you get the chance? We can get this into main.

@rustbot author

[asomers]

Rebased, @tgross35

[asomers]

I think the build failure on s390x is unrelated.

[tgross35]

That job seems to fail a lot, I wrote up #4112.

Thanks for the changes, I'll take a look today.

[tgross35]

I went through line by line, everything looks correct here but I left a handful of consistency comments.

[bors]

☔ The latest upstream changes (presumably #3978) made this pull request unmergeable. Please resolve the merge conflicts.

[asomers]

I applied your suggestions, @tgross35 , and rebased. But I did not squash the final commit.

[tgross35]

Does #3630 (comment) work? If not, a FIXME(ctest) like the other expressions would be good.

I'll merge tomorrow either way, please squash when you get the chance.

@rustbot label stable-declined (breakage)

[asomers]

Does #3630 (comment) work? If not, a FIXME(ctest) like the other expressions would be good.

I'll merge tomorrow either way, please squash when you get the chance.

@rustbot label stable-declined (breakage)

Whoops. I missed that comment. I've fixed it now.

[tgross35]

Thanks!