Use trait for labels instead of TypeId

[MichaReiser]

Summary

Fixes #5263

Rust 1.72.0 changes TypeId from a u64 to a u128. This breaks Ruffs release build because it increases the size of FormatElement.

This PR removes the dependency on TypeId for LabelId and instead introduces a new trait that downstream crates should implement on an enum defining all labels for that project.

This approach has two downsides:

• The API cannot statically enforce that a project uses a single LabelDefinition. Mixing two LabelDefinition could result in two labels being equal that should not be equal. I added a
  debug assertion that compares the names of the labels to mitigate the risk.
• Implementing the first Label requires more boilerplate code than before

One main upside is that it allows a project to define all labels in a single place (and search for all label usages).

Test Plan

• cargo test
• cargo build -p ruff_python_formatter --release on Rust 1.72.0

[MichaReiser]

Current dependencies on/for this PR:

• main
  • PR Use trait for labels instead of TypeId #5270 👈

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                 1.01      7.0±0.02ms     5.8 MB/sec    1.00      6.9±0.01ms     5.9 MB/sec
formatter/numpy/ctypeslib.py               1.04   1434.0±3.37µs    11.6 MB/sec    1.00   1385.1±1.99µs    12.0 MB/sec
formatter/numpy/globals.py                 1.01    134.0±0.24µs    22.0 MB/sec    1.00    133.3±1.19µs    22.1 MB/sec
formatter/pydantic/types.py                1.01      2.9±0.02ms     8.8 MB/sec    1.00      2.9±0.01ms     8.9 MB/sec
linter/all-rules/large/dataset.py          1.00     13.6±0.02ms     3.0 MB/sec    1.00     13.6±0.04ms     3.0 MB/sec
linter/all-rules/numpy/ctypeslib.py        1.00      3.5±0.00ms     4.8 MB/sec    1.00      3.5±0.00ms     4.8 MB/sec
linter/all-rules/numpy/globals.py          1.00    368.1±1.15µs     8.0 MB/sec    1.00    367.7±1.22µs     8.0 MB/sec
linter/all-rules/pydantic/types.py         1.00      6.0±0.01ms     4.2 MB/sec    1.00      6.0±0.02ms     4.2 MB/sec
linter/default-rules/large/dataset.py      1.01      7.0±0.02ms     5.8 MB/sec    1.00      7.0±0.01ms     5.8 MB/sec
linter/default-rules/numpy/ctypeslib.py    1.00   1474.0±3.81µs    11.3 MB/sec    1.00   1468.6±1.77µs    11.3 MB/sec
linter/default-rules/numpy/globals.py      1.00    154.6±0.22µs    19.1 MB/sec    1.00    154.3±0.32µs    19.1 MB/sec
linter/default-rules/pydantic/types.py     1.00      3.2±0.01ms     8.0 MB/sec    1.00      3.2±0.00ms     8.0 MB/sec

Windows

group                                      main                                   pr
-----                                      ----                                   --
formatter/large/dataset.py                 1.00      5.7±0.04ms     7.2 MB/sec    1.00      5.7±0.06ms     7.2 MB/sec
formatter/numpy/ctypeslib.py               1.04   1157.8±9.47µs    14.4 MB/sec    1.00  1118.3±32.14µs    14.9 MB/sec
formatter/numpy/globals.py                 1.01    110.0±2.37µs    26.8 MB/sec    1.00    109.4±2.03µs    27.0 MB/sec
formatter/pydantic/types.py                1.00      2.4±0.04ms    10.7 MB/sec    1.01      2.4±0.02ms    10.6 MB/sec
linter/all-rules/large/dataset.py          1.04     11.6±0.37ms     3.5 MB/sec    1.00     11.1±0.18ms     3.7 MB/sec
linter/all-rules/numpy/ctypeslib.py        1.05      3.1±0.08ms     5.3 MB/sec    1.00      3.0±0.03ms     5.6 MB/sec
linter/all-rules/numpy/globals.py          1.04   324.2±13.45µs     9.1 MB/sec    1.00    311.6±4.24µs     9.5 MB/sec
linter/all-rules/pydantic/types.py         1.04      5.2±0.18ms     4.9 MB/sec    1.00      5.0±0.06ms     5.1 MB/sec
linter/default-rules/large/dataset.py      1.00      5.9±0.18ms     6.9 MB/sec    1.00      5.9±0.10ms     7.0 MB/sec
linter/default-rules/numpy/ctypeslib.py    1.00  1214.2±12.30µs    13.7 MB/sec    1.00  1216.8±19.48µs    13.7 MB/sec
linter/default-rules/numpy/globals.py      1.01    130.4±2.17µs    22.6 MB/sec    1.00    128.7±1.78µs    22.9 MB/sec
linter/default-rules/pydantic/types.py     1.01      2.7±0.03ms     9.6 MB/sec    1.00      2.6±0.02ms     9.7 MB/sec