gh-150226: Speed up uuid.UUID string parsing

[esadomer]

gh-150226

Summary

This speeds up uuid.UUID construction from the documented string forms by handling the documented urn:uuid: prefix directly with str.removeprefix() before stripping braces and hyphens.

The previous broader normalization path is still used as a fallback whenever the direct normalization does not produce 32 hex characters, preserving legacy accepted inputs such as urn:<hex>.

Benchmark

Same-process benchmark on a local Windows CPython debug build, comparing origin/main:Lib/uuid.py with this patch while blocking _uuid to isolate the pure-Python constructor path:

plain_hex:  11584.5 ns -> 11366.8 ns (1.9%)
canonical:  12424.6 ns -> 12253.1 ns (1.4%)
braced:     13293.5 ns -> 12981.1 ns (2.3%)
urn:        13800.0 ns -> 12912.8 ns (6.4%)

Absolute timings are from a debug build; the benchmark compares old and new code in the same process.

Tests

• PCbuild\build.bat -d -p x64
• PCbuild\amd64\python_d.exe -m test test_uuid -v

I also attempted a local Windows Release build, but MSVC failed with an internal compiler error while compiling/linking externals\zlib-ng-2.2.4\arch\x86\chunkset_avx2.c, unrelated to this uuid change.

[python-cla-bot]

All commit authors signed the Contributor License Agreement.

[picnixz]

See issue for the rationale of closing this.