Extend BestFitting with mode

[MichaReiser]

Summary

This PR brings back the best fitting mode first introduced by #5184 and later reverted by 9a8ba58 because it was no longer necessary for the binary expression formatting.

This PR re-introduces the BestFitting mode to implement Black's behavior to only parenthesize non-breakable expressions if it makes them fit #6271.

x = aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
x_aaaa = aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa

# Output
x = aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
x_aaaa = (
	aaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaaa
)

This layout needs Best Fitting because with the new mode because

• it has three variations (not just two): flat, with parentheses, and flat as fallback
• Best fitting by default assumes that the content fits when reaching the first line break. This doesn't work for this use case because we need to measure if all content up to the ) fits (this is kind of the point)

Test Plan

Re-added tests. Used by the stacked PRs

[MichaReiser]

Current dependencies on/for this PR:

• main
  • PR Extend BestFitting with mode #6814 👈
    • PR Fix Printer group modes #6815
      • PR Maybe parenthesize long constants and names #6816
        • PR Use BestFits for non-fluent attribute chains #6817
          • PR Add tab width option #6848
          • PR Printer: Slice based queue and stack #6819

This comment was auto-generated by Graphite.

[github-actions]

PR Check Results

Ecosystem

✅ ecosystem check detected no changes.

Benchmark

Linux

group                                      main                                   pr
-----                                      ----                                   --
formatter/large/dataset.py                 1.00      4.0±0.19ms    10.1 MB/sec    1.00      4.0±0.12ms    10.1 MB/sec
formatter/numpy/ctypeslib.py               1.00   835.9±40.45µs    19.9 MB/sec    1.02   854.6±59.01µs    19.5 MB/sec
formatter/numpy/globals.py                 1.01     83.7±3.42µs    35.2 MB/sec    1.00     82.5±3.50µs    35.8 MB/sec
formatter/pydantic/types.py                1.01  1662.0±83.73µs    15.3 MB/sec    1.00  1650.3±50.47µs    15.5 MB/sec
linter/all-rules/large/dataset.py          1.00     12.3±0.27ms     3.3 MB/sec    1.01     12.4±0.32ms     3.3 MB/sec
linter/all-rules/numpy/ctypeslib.py        1.00      3.3±0.10ms     5.1 MB/sec    1.00      3.2±0.06ms     5.1 MB/sec
linter/all-rules/numpy/globals.py          1.02   482.4±21.24µs     6.1 MB/sec    1.00   473.3±16.10µs     6.2 MB/sec
linter/all-rules/pydantic/types.py         1.01      6.5±0.19ms     3.9 MB/sec    1.00      6.4±0.15ms     4.0 MB/sec
linter/default-rules/large/dataset.py      1.01      6.6±0.13ms     6.2 MB/sec    1.00      6.5±0.15ms     6.2 MB/sec
linter/default-rules/numpy/ctypeslib.py    1.00  1434.9±53.55µs    11.6 MB/sec    1.00  1434.4±51.23µs    11.6 MB/sec
linter/default-rules/numpy/globals.py      1.00    174.2±6.46µs    16.9 MB/sec    1.02    177.0±9.12µs    16.7 MB/sec
linter/default-rules/pydantic/types.py     1.02      3.0±0.09ms     8.6 MB/sec    1.00      2.9±0.07ms     8.7 MB/sec

Windows

group                                      main                                   pr
-----                                      ----                                   --
formatter/large/dataset.py                 1.01      3.0±0.04ms    13.4 MB/sec    1.00      3.0±0.02ms    13.5 MB/sec
formatter/numpy/ctypeslib.py               1.00    574.8±8.43µs    29.0 MB/sec    1.01    578.6±8.48µs    28.8 MB/sec
formatter/numpy/globals.py                 1.00     58.0±0.78µs    50.9 MB/sec    1.02     59.3±1.27µs    49.7 MB/sec
formatter/pydantic/types.py                1.00  1206.5±16.21µs    21.1 MB/sec    1.00  1202.1±15.30µs    21.2 MB/sec
linter/all-rules/large/dataset.py          1.00      9.4±0.19ms     4.3 MB/sec    1.02      9.6±0.10ms     4.3 MB/sec
linter/all-rules/numpy/ctypeslib.py        1.00      2.6±0.06ms     6.3 MB/sec    1.02      2.7±0.04ms     6.2 MB/sec
linter/all-rules/numpy/globals.py          1.00    272.1±7.74µs    10.8 MB/sec    1.02    277.9±9.48µs    10.6 MB/sec
linter/all-rules/pydantic/types.py         1.00      4.9±0.17ms     5.2 MB/sec    1.01      5.0±0.13ms     5.1 MB/sec
linter/default-rules/large/dataset.py      1.00      5.2±0.09ms     7.8 MB/sec    1.01      5.2±0.06ms     7.8 MB/sec
linter/default-rules/numpy/ctypeslib.py    1.00  1073.7±17.95µs    15.5 MB/sec    1.02  1100.2±21.05µs    15.1 MB/sec
linter/default-rules/numpy/globals.py      1.00    108.5±1.53µs    27.2 MB/sec    1.00    108.8±1.64µs    27.1 MB/sec
linter/default-rules/pydantic/types.py     1.00      2.4±0.03ms    10.7 MB/sec    1.00      2.4±0.03ms    10.7 MB/sec

[MichaReiser]

I got very flacky results without this.

[MichaReiser]

I'll merge this without review. You already reviewed the original PR