Remove AttributeExt

[mejrs]

Part of #131229 (comment)

Removes AttributeExt, turning some of its methods into inherent methods where necessary.

I had to choose between fn is_doc_comment(&self) -> Option<Span> (trait) and fn is_doc_comment(&self) -> bool (inherent). I chose the latter because nothing is actually using the span.

This PR does not make further changes as

• that changes many files and would be very prone to merge conflicts
• many places that call methods like path and name are subtly wrong and IMO should be fixed in standalone prs.

[rustbot]

Some changes occurred in compiler/rustc_attr_parsing

cc @jdonszelmann, @JonathanBrouwer

[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

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler, parser
• compiler, parser expanded to 69 candidates
• Random selection from 12 candidates

[mejrs]

View changes since the review

This is an example of one of those "subtly wrong" places. #154924 will take care of it.

[rust-log-analyzer]

The job pr-check-2 failed! Check out the build log: (web) (plain enhanced) (plain)

Click to see the possible cause of the failure (guessed by this bot)

     | ^^^^^^^^^^^^^^^^^^^^
note: function defined here
    --> compiler/rustc_resolve/src/rustdoc.rs:358:8
     |
 358 | pub fn inner_docs(attrs: &[ast::Attribute]) -> bool {
     |        ^^^^^^^^^^

error[E0271]: expected `impl Iterator<Item = (&'a hir::Attribute, Option<DefId>)>` to be an iterator that yields `(&Attribute, Option<DefId>)`, but it yields `(&Attribute, Option<DefId>)`
    --> src/librustdoc/clean/types.rs:1066:65
     |
1066 |         let (doc_strings, other_attrs) = attrs_to_doc_fragments(attrs, doc_only);
     |                                          ---------------------- ^^^^^ expected `rustc_ast::Attribute`, found `rustc_hir::Attribute`
     |                                          |
     |                                          required by a bound introduced by this call
     |
     = note: `rustc_hir::Attribute` and `rustc_ast::Attribute` have similar names, but are actually distinct types
---
    --> compiler/rustc_resolve/src/rustdoc.rs:200:26
     |
 199 | pub fn attrs_to_doc_fragments<'a>(
     |        ---------------------- required by a bound in this function
 200 |     attrs: impl Iterator<Item = (&'a ast::Attribute, Option<DefId>)>,
     |                          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ required by this bound in `attrs_to_doc_fragments`

error[E0308]: mismatched types
    --> src/librustdoc/clean/types.rs:1067:35
     |
1067 |         Attributes { doc_strings, other_attrs }
---

error[E0308]: mismatched types
    --> src/librustdoc/passes/collect_intra_doc_links.rs:1077:64
     |
1077 |                 && !has_primitive_or_keyword_or_attribute_docs(&item.attrs.other_attrs)
     |                     ------------------------------------------ ^^^^^^^^^^^^^^^^^^^^^^^ expected `&[Attribute]`, found `&ThinVec<Attribute>`
     |                     |
     |                     arguments to this function are incorrect
     |
     = note: expected reference `&[rustc_ast::Attribute]`
                found reference `&ThinVec<rustc_hir::Attribute>`
note: function defined here
    --> compiler/rustc_resolve/src/rustdoc.rs:368:8
     |
 368 | pub fn has_primitive_or_keyword_or_attribute_docs(attrs: &[ast::Attribute]) -> bool {
     |        ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

Some errors have detailed explanations: E0271, E0308, E0432.
For more information about an error, try `rustc --explain E0271`.
[RUSTC-TIMING] rustdoc test:false 3.454

[JonathanBrouwer]

We need to wait with merging this until we've made a decision on whether we revert the lint attribut port (see #152369), I'll review when the decision has been made