Enhance macro_use_imports lint with a auto applicable suggstion

[flip1995]

The macro_use_imports lint was implemented in #5230, but is currently missing a auto applicable suggestion. For this suggestion, every macro used in the module would have been detected and from this set of macros, a suggestion has to be build.

[DevinR528]

I have made a rough start on this here. I had a few questions.
What do we do with re exported paths such as println from std::prelude the macro is from std::macros which is private? this isn't a great example since not many would explicitly import std::prelude but I'd imagine it holds true for other crates

If the import is only the crate name (no more ::mod::mod) because macros float to the top level is it safe to assume crate::macro` is always valid?

I'm sure there are other things I'm not doing correctly, should I move this to a PR?

[flip1995]

I must admit, that I don't really know, how I would handle this.

I think that crate::macro can still be a reexport from a private module of crate. It will definitely need extensive testing, including macros from sub crates, similar to

rust-clippy/tests/ui/wildcard_imports.rs

Lines 1 to 8 in 3d0f0e3

	// run-rustfix
	// aux-build:wildcard_imports_helper.rs
	#![warn(clippy::wildcard_imports)]
	#![allow(unused)]
	#![warn(unused_imports)]
	extern crate wildcard_imports_helper;

What I would suggest is to first write a huge test base, with every case you can think of, and then work from there. If you have an idea for a test case, but don't know how to implement it in the Clippy test suite, you can open a PR, and we can work it out together.