… r=wesleywiser

Always register sized obligation for argument

Removes a "hack" that skips registering sized obligations for parameters that are simple identifiers. This doesn't seem to affect diagnostics because we're probably already being smart enough about deduplicating identical error messages anyways.

Fixes rust-lang#112608

…ter, r=notriddle

Add link to rustdoc book search chapter in help popover

One thing that was missing in the rustdoc output and its help popover was a link back to the search feature chapter in the rustdoc book.

It looks like this:

![image](https://github.com/rust-lang/rust/assets/3050060/425413e5-e734-4d40-b675-8b2dcef874a2)

r? ```@notriddle```

…yout, r=wesleywiser

Don't ICE on unnormalized struct tail in layout computation

1. We try to compute a `SizeSkeleton` even if a layout error occurs, but we really only need to do this if we get `LayoutError::Unknown`, since that means our type is too polymorphic to actually compute the full layout. If we have other errors, like `LayoutError::NormalizationError` or `LayoutError::Cycle`, then we can't really make any progress, since this represents an actual error.
2. Avoid using `normalize_erasing_regions` and `struct_tail_erasing_lifetimes` since those ICE on normalization errors, and since we may call `layout_of` in HIR typeck, we don't know for certain that we're on the happy path.

Fixes rust-lang#112736

Stop hiding const eval limit in external macros

fixes rust-lang#112748

We don't emit a hard error if there was a previous deny lint triggering with the same message. If that lint ends up not being emitted, we ICE and don't emit an error either.

…, r=notriddle

[tests/rustdoc] Add @files command

The ````@!has```` checks is very problematic as it wouldn't catch if the file scheme is updated and the file is generated again. ````@files```` allows to ensure that the given folder contains exactly the provided entries (files and folders).

I'm wondering if we should forbid the ````@!has```` for files. To be discussed after this PR I suppose.

r? ```@notriddle```

…ter, r=ozkanonur

Fix rustdoc gui tester

Problem was that the `main` was always exiting with `0`, whether or not there was an error. It led to failing GUI tests being ignored in the CI since no one saw them.

r? ``@klensy``