feat(expr): add timestamp timezone replacement

[danking]

Timestamp expressions need to reinterpret local wall times in another timezone, including daylight-saving transitions. Add native replace_time_zone to Vortex's Rust and Python expression APIs, preserving timestamp units and propagating nulls.

Support timezone removal, per-row raise/earliest/latest/null ambiguity policies, and raise/null policies for nonexistent times. Serialize options through protobuf and register the function for expression deserialization. Polars expression mapping is in the stacked datetime PR #10183.

Validation: regenerated bindings with cargo run -p xtask -- generate-proto. Added Rust cases for timestamp units, DST folds/gaps, overflow, invalid inputs, and option serialization, plus Python constant/null cases. Tests and linting were not run.

[danking]

I need to look at this carefully

[codspeed]

Merging this PR will degrade performance by 5.73%

⚠️ 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.

⚡ 3 improved benchmarks
❌ 7 regressed benchmarks
✅ 2165 untouched benchmarks
🆕 2 new benchmarks
⏩ 434 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	WallTime	mul_u64_nonnull_neon	28.5 µs	39.6 µs	-27.91%
❌	WallTime	multiply_shapes_neon[(32768, PerRowPerRow)]	32.3 µs	38.8 µs	-16.86%
❌	WallTime	mul_i64_nonnull_neon	32.6 µs	38.7 µs	-15.86%
❌	Simulation	sum_i64	193.9 µs	224.9 µs	-13.76%
❌	Simulation	sum_v2_i64	193.8 µs	224.4 µs	-13.61%
❌	WallTime	scalar_subtract_neon	10.8 µs	12.4 µs	-13.16%
❌	WallTime	bitpack_blocked_compress_avx2	6.8 µs	7.6 µs	-11.13%
⚡	Simulation	decode_primitives[i64, (1000, 32)]	37.7 µs	24.5 µs	+53.82%
⚡	WallTime	compare_u64_avx2	3.8 µs	3.4 µs	+11.69%
⚡	Simulation	or_true_constant	20.3 µs	18.2 µs	+11.31%
🆕	Simulation	varbinview_fill_null_inlined	N/A	276.8 µs	N/A
🆕	Simulation	varbinview_fill_null_outlined	N/A	289.1 µs	N/A

Tip

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

---

_{Comparing dk/replace-time-zone (92ddab2) with develop (53c7659)²}

Footnotes

1. 434 benchmarks were skipped, so the baseline results were used instead. If they were deleted from the codebase, click here and archive them to remove them from the performance reports. ↩

2. No successful run was found on develop (30e84dc) during the generation of this report, so 53c7659 was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩

[robert3005]

I have applied a couple of fixes here. This should be good now