Fix Validity::mask_eq semantics for mixed variants

[joseph-isaacs]

Summary

PR 2 of a 4-PR stack (stacked on #8333) preparing Validity for lazy validity arrays.

mask_eq previously returned false for any mixed-variant pairing without executing — e.g. a Validity::Array that resolves to all-true compared against Validity::AllValid. With lazy validity arrays, unresolved Array variants frequently hold constant masks, making this silently wrong rather than merely conservative.

• mask_eq now takes the validity length and compares logical masks: constant variants keep their no-execute fast paths, and any pairing involving an Array executes both sides via execute_mask and compares the resulting Masks (whose PartialEq is already representation-independent).
• NonNullable and AllValid now compare equal, since their logical masks are identical.
• Adds 9 rstest cases covering mixed constant pairings and arrays resolving to constant masks.

All existing callers (tests/asserts in vortex, datetime-parts, pco, zstd, builders, masked) are updated to pass the length.

Checks

• cargo nextest run -p vortex-array (2962 passed, incl. new mask_eq_mixed_variants cases)
• cargo nextest run -p vortex-datetime-parts -p vortex-pco -p vortex-zstd (119 passed)
• cargo build --tests -p vortex, cargo clippy -p vortex-array --all-targets, cargo +nightly fmt --all

https://claude.ai/code/session_01VPQ7dfZtijfrsjAipwXvEj

---

Generated by Claude Code

[codspeed-hq]

Merging this PR will degrade performance by 18.29%

⚠️ Unknown Walltime execution environment detected

Using the Walltime instrument on standard Hosted Runners will lead to inconsistent data.

For the most accurate results, we recommend using CodSpeed Macro Runners: bare-metal machines fine-tuned for performance measurement consistency.

❌ 5 regressed benchmarks
✅ 1521 untouched benchmarks

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	chunked_bool_canonical_into[(1000, 10)]	31.8 µs	47 µs	-32.34%
❌	Simulation	chunked_varbinview_canonical_into[(1000, 10)]	161.8 µs	198.6 µs	-18.5%
❌	Simulation	chunked_varbinview_into_canonical[(1000, 10)]	177.2 µs	213.2 µs	-16.85%
❌	Simulation	chunked_varbinview_canonical_into[(100, 100)]	274.5 µs	308.9 µs	-11.15%
❌	Simulation	bitwise_not_vortex_buffer_mut[128]	246.1 ns	275.3 ns	-10.6%

Tip

Investigate this regression by commenting @codspeedbot fix this regression on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing claude/cool-bardeen-l8jlsy-2-mask-eq (4813bec) with claude/cool-bardeen-l8jlsy-1-definitely-no-nulls (4906359)}