feat(linter): implement noUselessStringRaw

[fireairforce]

Summary

closes: #3945

Test Plan

[codspeed-hq]

CodSpeed Performance Report

Merging #4263 will improve performances by 8.53%

_{Comparing fireairforce:feat-3945 (3dd6827) with main (d254b97)}

Summary

⚡ 1 improvements
✅ 104 untouched benchmarks

Benchmarks breakdown

	Benchmark	main	fireairforce:feat-3945	Change
⚡	db_17847247775464589309.json[cached]	13.9 ms	12.8 ms	+8.53%

[Conaclos]

Calling text() on a node allocates a string.
We could avoid it by retrieving the token and the calling text_trimmed_text()

Suggested change

	let object_name = object_expr.name().ok()?.text();
	let object_name = object_expr.name().ok()?.names().ok()?.value_token().ok()?;
	let object_name = object_name.text_trimmed_text();

[fireairforce]

updated~ maybe you mean text_trimmed() here.

[Conaclos]

Suggested change

	let member_name = member.as_js_name()?.text();
	let member_name = member.as_js_name()?.value_token().ok()?;
	let member_name = member_name.text_trimmed_text();

[fireairforce]

updated~

[Conaclos]

Iterating over chars() is expensive because this decodes UTF-8 characters.
Here we could directly operate over bytes.
Also, peekable() has some overhead that can be often avoided.
Here is my suggestion:

Suggested change

	let mut chars = input.chars().peekable();
	while let Some(ch) = chars.next() {
	if ch == '\\' {
	if let Some('n' | 't' | 'r' | '\\' | '"' | '\'' | '0') = chars.peek() {
	return true;
	}
	}
	}
	false
	let mut bytes = input.bytes();
	while let Some(c) = bytes.next() {
	if c == b'\\' {
	if matches!(bytes.next(), Some(b'n' | b't' | b'r' | b'\\' | b'"' | b'\'' | b'0')) {
	return true;
	}
	}
	}
	false

Actually I wonder even if we could just check for \ because it will be directly interpreted in String.raw and regular strings. For instance in a regular string \a is a useless escape, however it will be results in the string a. While in a raw string this results in the string \a.
Thus, we could just check if the string contains \. What do you think?

    input.contains('\\')

[fireairforce]

yes, i agree you, It is logically more reasonable to only check \

[Conaclos]

Looks good to me! Thanks! Could you add a changelog entry?

[fireairforce]

it seems i have added the changelog here, what else i need to add ? @Conaclos

[Conaclos]

Sorry I missed it. All is perfect :)