Support self at the end of any paths in imports

[mu001999]

As a follow-up PR to #146972, after this PR:

1. self can appear at the end of any paths in imports (consensus in Support importing path-segment keyword with renaming #146972 (comment))
2. E0429 will be no longer emitted, and new lint redundant_self will be emitted for use ...::self [as target]; or use ...::{self [as target]};

For the second, the new lint and clippy lint unnecessary_self_imports have some overlap and also cause the following known issues:

Removing ::{self} will cause any non-module items at the same path to also be imported. This might cause a naming conflict (rust-lang/rustfmt#3568).

So perhaps this new lint should not be added but extend the clippy lint unnecessary_self_imports?

r? petrochenkov

[petrochenkov]

It looks like #146972 (comment) suggests to support trailing self in all paths, not just import paths. We can start with import paths if that's more convenient though.

However, if we are doing it, then I think it's time to abandon the whole "self rewriting" hack, and just properly resolve self inside modules.

[mu001999]

It looks like #146972 (comment) suggests to support trailing self in all paths, not just import paths.

Semantics of trailing self in import paths, like use foo::self;, is explicit and same to use foo::{self};, and this will require foo is in type namespace. So I think trailing self in import paths is usefull to replace use foo::{self}; when we want to import only items in type namespace.

But what will trailing self in non-import paths be? Do we need things like Foo::self? 🤔

[petrochenkov]

Same as with imports, for foo::self to resolve foo must be a module in name resolution sense (i.e. module, enum or trait), otherwise it's a resolution error.

[petrochenkov]

Ugh, apparently things like this work.

struct S {}

use S::{self as other};

This isn't supposed to happen.

[mu001999]

Same as with imports, for foo::self to resolve foo must be a module in name resolution sense (i.e. module, enum or trait), otherwise it's a resolution error.

What I mean is that trailing self in imports can be used to restrict the resolve behavior, but what can we use trailing self in non-import paths to do? 🤔 Maybe pub(crate::x::foo::self), but do we really need such things?

[mu001999]

Ugh, apparently things like this work.

struct S {}

use S::{self as other};

This isn't supposed to happen.

😂 Looks like we need module_only rather than type_ns_only when importing