restore TyCtxt::push_impl_path() logic for checking whether types are available

[dwrensha]

Fixes #41697 by restoring some logic that got removed in 2cca256.

Note that I don't have a super strong grasp of what exactly is happening is this code. It's entirely possible that @nikomatsakis had a good reason for removing these lines, in which case maybe there is a better fix.

[rust-highfive]

Thanks for the pull request, and welcome! The Rust team is excited to review your changes, and you should hear from @nikomatsakis (or someone else) soon.

If any changes to this PR are deemed necessary, please add them as extra commits. This ensures that the reviewer can see what has changed since they last reviewed the code. Due to the way GitHub handles out-of-date commits, this should also make it reasonably obvious what issues have or haven't been addressed. Large or tricky changes may require several passes of review and changes.

Please see the contribution instructions for more information.

[Mark-Simulacrum]

Thanks for the PR! We'll periodically check in to make sure that @nikomatsakis or someone else from the team reviews this soon.

[oli-obk]

This should probably have a regression test so it doesn't regress again. Adding

// dump-mir as regression test for #41697
// compile-flags:-Zdump-mir=all

to src/test/run-pass/issue-33387.rs should to it.

[dwrensha]

@oli-obk If I set that flag (on issue-33387.rs or issue-41697.rs) then running the test dumps a bunch of files like rustc.node15.000-000.SimplifyCfg-initial.after.mir in the project root directory. Maybe we could avoid that by also adding a -Zdump-mir-dir=DIR flag, but I'm unsure what directory would make sense to put there.

[nikomatsakis]

Sadly, this fix isn't quite right. We shouldn't be poking directly at the fields of the maps structure (and, in fact, I'm surprised this even compiles, since they are supposed to be private). I have a (midly) better fix for #41697 specifically, though I think the best fix is to rewrite the item-path code more thoroughly. I was trying to decide if I could come up with a better unit test though and am not sure how best to do it.

[nikomatsakis]

I'm going to close in favor of #42017 -- thanks @dwrensha !