Draft: 12015 - suggest const thread_local

[m-rph]

Currently Draft.

The lint aims to suggest using a const initializer in thread_local calls if possible:

thread_local! {
   static ... = const { RefCell::new(0) };
}

if possible.

After expanding the macro, I observed that if a const is given, there is no __init fn. Thus, we can check the init fn and it's actual body for the initializer value. This is currently in draft because I am not sure how to correctly check whether the expression can be evaluated as a const, and I hit an ICE with the two functions I tried to use.

Attempts to close: #12015

changelog: New Lint [suggest_const_thread_local]: suggest const { initializer } for threadlocal if possible.

• Followed [lint naming conventions][lint_naming]
• Added passing UI tests (including committed .stderr file)
• cargo test passes locally
• Executed cargo dev update_lints
• Added lint documentation
• Run cargo dev fmt

[rustbot]

r? @Jarcho

(rustbot has picked a reviewer for you, use r? to override)