Support type aliases in union passthrough

[yzhkali]

Summary

• unwrap PEP 695 type aliases when union passthrough checks native union members
• keep spillover handling for non-passthrough union members intact
• add Python 3.12 coverage for a type alias mixed with a dataclass in the target union

Closes #742

Tests

• .venv/bin/python -m pytest tests/strategies/test_tagged_unions_695.py tests/strategies/test_union_passthrough.py tests/strategies/test_union_passthrough_695.py
• .venv/bin/ruff check src/cattrs/strategies/_unions.py tests/strategies/test_union_passthrough_695.py
• .venv/bin/ruff format --check src/cattrs/strategies/_unions.py tests/strategies/test_union_passthrough_695.py
• git diff --check

[codspeed-hq]

Merging this PR will improve performance by 21.14%

⚠️ Different runtime environments detected

Some benchmarks with significant performance changes were compared across different runtime environments,
which may affect the accuracy of the results.

Open the report in CodSpeed to investigate

#### 🎉 Hooray! `pytest-codspeed` just leveled up to 5.0.3!

A heads-up, this is a breaking change and it might affect your current performance baseline a bit. But here's the exciting part - it's packed with new, cool features and promises improved result stability 🥳!
Curious about what's new? Visit our releases page to delve into all the awesome details about this new version.

⚡ 38 improved benchmarks
✅ 26 untouched benchmarks

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
⚡	test_unstructure_hetero_tuple[BaseConverter]	14.9 µs	11 µs	+35.42%
⚡	test_unstructure_int[BaseConverter]	15.5 µs	11.9 µs	+29.98%
⚡	test_unstructure_simple_str_enum[MsgspecJsonConverter]	16.2 µs	12.4 µs	+29.93%
⚡	test_unstructure_float[Converter]	16.2 µs	12.5 µs	+29.52%
⚡	test_unstructure_simple_str_enum[BaseConverter]	15.5 µs	12 µs	+29.28%
⚡	test_unstructure_simple_str_enum[Converter]	16.1 µs	12.4 µs	+29.22%
⚡	test_unstructure_float[BaseConverter]	15.5 µs	12 µs	+29.12%
⚡	test_unstructure_int[Converter]	16.1 µs	12.5 µs	+29.11%
⚡	test_unstructure_simple_str_enum[OrjsonConverter]	16.2 µs	12.6 µs	+27.82%
⚡	test_unstructure_hetero_tuple[Converter]	17.4 µs	13.8 µs	+26.31%
⚡	test_unstructure_simple_enum[BaseConverter]	20.6 µs	16.4 µs	+25.15%
⚡	test_unstructure_simple_enum[Converter]	19.4 µs	15.5 µs	+25.13%
⚡	test_unstructure_simple_int_enum[Converter]	19.9 µs	16 µs	+24.76%
⚡	test_unstructure_simple_enum[MsgspecJsonConverter]	19.4 µs	15.6 µs	+24.75%
⚡	test_unstructure_simple_int_enum[OrjsonConverter]	19.7 µs	15.8 µs	+24.66%
⚡	test_unstructure_simple_int_enum[MsgspecJsonConverter]	19.7 µs	15.8 µs	+24.66%
⚡	test_unstructure_simple_enum[OrjsonConverter]	19.3 µs	15.6 µs	+23.49%
⚡	test_unstructure_simple_int_enum[BaseConverter]	20.9 µs	17.1 µs	+22.19%
⚡	test_structure_simple_str_enum[BaseConverter]	21.5 µs	17.9 µs	+20.2%
⚡	test_structure_simple_enum[Converter]	22.5 µs	18.8 µs	+20.01%
...	...	...	...	...

ℹ️ Only the first 20 benchmarks are displayed. Go to the app to view all benchmarks.

Tip

Curious why this is faster? Comment @codspeedbot explain why this is faster on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing yzhkali:fix/union-passthrough-type-alias (81a1dc6) with main (087e1ce)¹}

Footnotes

1. No successful run was found on main (a735ca6) during the generation of this report, so 087e1ce was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[Tinche]

Looks ok to me. Please add a changelog entry, and versionchanged blocks to both the function and the docs.

The test failure looks like a GitHub issue?