perf(row): compact variable-list keys

[gatesn]

Summary

• replace per-byte list escaping with one element marker followed by the complete child row key
• retain list ranges, canonical children, child sizes, and offsets from the sizing pass for encoding
• preserve configurable nested null ordering with a dedicated element marker
• add a regression test for a null child sentinel colliding with the ascending list terminator

Depends on #9553.

Performance

cargo bench -p vortex-row --bench row_encode -- --output-sizes

Case	Before	After	Reduction
List<i64>	146,000 bytes	82,000 bytes	43.8%
List<Utf8>	546,000 bytes	282,000 bytes	48.4%

Local median encode times with 1,000 lists of 8 elements:

Case	Before	After	Improvement
List<i64>	46.74 us	28.20 us	39.7%
List<Utf8>	100.7 us	98.08 us	2.6%

The corresponding arrow-row medians in the before/after runs were 36.83/36.56 us for List<i64> and 94.74/97.62 us for List<Utf8>, providing a noise anchor.

Testing

• cargo test -p vortex-row (33 tests plus doctest)
• cargo clippy -p vortex-row --all-targets --all-features
• cargo +nightly fmt --all -- --check
• git diff --check
• cargo bench -p vortex-row --bench row_encode -- --output-sizes
• cargo bench -p vortex-row --bench row_encode -- list_ --sample-count 20 --min-time 0.2 --max-time 0.5 --color never

[codspeed-hq]

Merging this PR will improve performance by 25.68%

⚠️ 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.

⚠️ 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

⚡ 2 improved benchmarks
✅ 1983 untouched benchmarks
⏩ 54 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Simulation	list_i64_vortex	2.2 ms	1.6 ms	+34.55%
⚡	Simulation	list_utf8_vortex	4.1 ms	3.5 ms	+17.39%

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 ngates/vortex-row-list-performance (87996f3) with ngates/vortex-row-dtype-support (41da939)}

Footnotes

1. 54 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. ↩