perf: branchless mask-select for listview zip

[joseph-isaacs]

Summary

Replaces the per-element, data-dependent branch in the listview zip kernel's offset/size selection with a branchless, chunk-at-a-time mask select that the compiler can auto-vectorize.

For each 64-bit mask chunk, each bit is expanded to a full-width lane mask and both sides are blended with (t & m) | (f & !m) via a shared select_column helper, so the inner loop is branch-free regardless of mask shape. if_false offsets are shifted into the second half of the concatenated elements as before; sizes are taken verbatim from the chosen side.

The primitive-array zip kernel that was previously bundled here has been split into its own PR.

Changes

• vortex-array/src/arrays/listview/compute/zip.rs: branchless chunked offset/size select via select_column; added a test that spans the 64-bit chunk boundary + remainder.
• vortex-array/benches/listview_zip.rs: a small divan bench — one Fragmented (alternating) mask with non-nullable and nullable inputs.

Benchmark

Because the branchless select is mask-shape-independent, the bench uses a single Fragmented mask (the worst case for the replaced per-element branch) with both nullability variants. LEN is 8,192: listview zip cost is dominated by element concatenation and per-list canonicalization, so a few thousand lists already exercise the select while keeping each case well under a few hundred microseconds (~37 µs non-null, ~41 µs nullable locally).

The kernel change itself is a single-to-low-double-digit-percent win on the offset/size select; end-to-end listview zip is dominated by element concatenation + canonicalization, so the select is a small slice of total cost.

Testing

• vortex-array listview zip tests pass (incl. the new multi-chunk-span test); cargo +nightly fmt and clippy -D warnings (default + all-features) clean.

https://claude.ai/code/session_01N5ivPiCJy7dGQjMEP7ips9

[codspeed-hq]

Merging this PR will not alter performance

⚠️ 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.

✅ 1595 untouched benchmarks
🆕 2 new benchmarks
⏩ 4 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
🆕	Simulation	nonnull	N/A	302.5 µs	N/A
🆕	Simulation	nullable	N/A	315.4 µs	N/A

---

_{Comparing claude/wizardly-carson-6Cixf (3b5db68) with develop (0026d3a)}

Footnotes

1. 4 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. ↩

[github-actions]

This PR has been marked as stale because it has been open for 14 days with no activity. Please comment or remove the stale label if you wish to keep it active, otherwise it will be closed in 7 days