refactor: pack and gather patches with explicit chunk widths

[mhk197]

Share packing and exception gathering between uniform and explicit per-chunk layouts. Expose an encoder for caller-supplied chunk widths, including full-width and zero-width chunks, while retaining the global-width entry points. Cover full-width chunks and zero-copy slicing with nonzero offset origins.

Part 5/9 of the bitpacked-v2 stack (545 added + removed lines). Merge in order; each draft targets the preceding branch, starting from develop.

1 → 2 → 3 → 4 → 5 → 6 → 7 → 8 → 9

Validation: All 13 focused plugin, serde, and chunk-layout tests passed on this commit, including the regression that rejects direct vtable serde. Changed Rust files were formatted with nightly; git diff --check passed.

[codspeed]

Merging this PR will degrade performance by 14.65%

⚠️ 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

⚡ 1 improved benchmark
❌ 5 regressed benchmarks
✅ 2247 untouched benchmarks
⏩ 218 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	decode_primitives[f32, (1000, 512)]	42 µs	64.3 µs	-34.67%
❌	Simulation	random_i8[0.5]	72.4 µs	96 µs	-24.6%
❌	Simulation	bitpacked_compress_u32	49.6 µs	60.7 µs	-18.29%
❌	WallTime	words_gather_scalar_avx2[65536]	8.3 µs	9.4 µs	-11.8%
❌	WallTime	mul_u32_nonnull_avx512	5.6 µs	6.3 µs	-11.01%
⚡	Simulation	random_i16[0.95]	99.7 µs	81.5 µs	+22.34%

Tip

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

---

_{Comparing mk/bitpacked-stack-05-explicit-packing (120d3bd) with mk/bitpacked-stack-04-offset-child (cfa9850)}

Footnotes

1. 218 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. ↩