Omit tuple parentheses inside comprehensions

[cnpryer]

Summary

Closes #5779

Implements TupleParentheses::Comprehension for omitting parentheses inside comprehensions.

Test Plan

Review updated snapshots.

---

I feel like this kind of thing can be implemented a number of ways. One way I was thinking about was a NeedsParentheses implementation. I imagined something like

impl NeedsParentheses for ExprTuple {
    fn needs_parentheses(
        &self,
        parent: AnyNodeRef,
        _context: &PyFormatContext,
    ) -> OptionalParentheses {
        if parent.is_comprehension() {
            ...
        }
    }
}

I could be mixing concepts though. I'm also wondering if this warrants just having TupleParentheses::Never.

TODO:

• Handle parenthesized Argument 3bc182b
• More fixtures

[github-actions]

PR Check Results

Ecosystem

✅ ecosystem check detected no changes.

Benchmark

Linux

group                                      main                                   pr
-----                                      ----                                   --
formatter/large/dataset.py                 1.00      9.8±0.05ms     4.2 MB/sec    1.01      9.8±0.04ms     4.1 MB/sec
formatter/numpy/ctypeslib.py               1.00   1904.1±5.72µs     8.7 MB/sec    1.00   1910.2±3.86µs     8.7 MB/sec
formatter/numpy/globals.py                 1.00    207.6±0.45µs    14.2 MB/sec    1.01    209.0±0.82µs    14.1 MB/sec
formatter/pydantic/types.py                1.00      4.2±0.01ms     6.1 MB/sec    1.00      4.2±0.01ms     6.1 MB/sec
linter/all-rules/large/dataset.py          1.00     13.6±0.11ms     3.0 MB/sec    1.00     13.5±0.08ms     3.0 MB/sec
linter/all-rules/numpy/ctypeslib.py        1.00      3.4±0.01ms     4.9 MB/sec    1.00      3.4±0.00ms     4.9 MB/sec
linter/all-rules/numpy/globals.py          1.00    373.9±0.76µs     7.9 MB/sec    1.00    373.0±2.76µs     7.9 MB/sec
linter/all-rules/pydantic/types.py         1.00      6.1±0.06ms     4.2 MB/sec    1.00      6.1±0.06ms     4.2 MB/sec
linter/default-rules/large/dataset.py      1.01      7.1±0.03ms     5.7 MB/sec    1.00      7.1±0.03ms     5.8 MB/sec
linter/default-rules/numpy/ctypeslib.py    1.00   1440.4±4.38µs    11.6 MB/sec    1.00   1434.0±5.05µs    11.6 MB/sec
linter/default-rules/numpy/globals.py      1.01    150.5±0.38µs    19.6 MB/sec    1.00    149.3±0.49µs    19.8 MB/sec
linter/default-rules/pydantic/types.py     1.00      3.2±0.02ms     8.1 MB/sec    1.00      3.2±0.01ms     8.1 MB/sec

Windows

group                                      main                                   pr
-----                                      ----                                   --
formatter/large/dataset.py                 1.00     10.3±0.18ms     3.9 MB/sec    1.01     10.4±0.25ms     3.9 MB/sec
formatter/numpy/ctypeslib.py               1.00  1936.9±26.87µs     8.6 MB/sec    1.01  1959.1±32.26µs     8.5 MB/sec
formatter/numpy/globals.py                 1.00    208.0±1.72µs    14.2 MB/sec    1.00    207.9±2.18µs    14.2 MB/sec
formatter/pydantic/types.py                1.00      4.3±0.07ms     5.9 MB/sec    1.01      4.3±0.08ms     5.9 MB/sec
linter/all-rules/large/dataset.py          1.00     15.0±0.48ms     2.7 MB/sec    1.02     15.3±0.43ms     2.7 MB/sec
linter/all-rules/numpy/ctypeslib.py        1.00      3.7±0.13ms     4.6 MB/sec    1.05      3.8±0.18ms     4.3 MB/sec
linter/all-rules/numpy/globals.py          1.00    399.3±5.48µs     7.4 MB/sec    1.00   399.4±11.91µs     7.4 MB/sec
linter/all-rules/pydantic/types.py         1.00      6.5±0.12ms     3.9 MB/sec    1.00      6.5±0.26ms     3.9 MB/sec
linter/default-rules/large/dataset.py      1.06      7.9±0.16ms     5.2 MB/sec    1.00      7.4±0.12ms     5.5 MB/sec
linter/default-rules/numpy/ctypeslib.py    1.04  1535.1±21.35µs    10.8 MB/sec    1.00  1471.2±15.37µs    11.3 MB/sec
linter/default-rules/numpy/globals.py      1.03    166.1±2.04µs    17.8 MB/sec    1.00    161.8±1.52µs    18.2 MB/sec
linter/default-rules/pydantic/types.py     1.04      3.3±0.03ms     7.7 MB/sec    1.00      3.2±0.04ms     8.0 MB/sec

[MichaReiser]

Nice. This looks good to me. My main question is about how this fits into your other work on parenthesizing tuples.