feat: select bitpacked widths independently per chunk

[mhk197]

Choose a cost-model width for each 1024-value chunk, accounting for padded packed bytes and exceptions. Add the per-chunk encoder with separate planning, packing, and patch-gathering passes. Temporary width choices produce one persistent offsets child.

Cover per-chunk decoding, zero-width chunks, nullable and signed values, and kernel conformance. The existing BtrBlocks scheme continues to use its global-width encoder.

Validation: 428 FastLanes/BtrBlocks tests passed (1 skipped).

[codspeed]

Merging this PR will regress 1 benchmark

⚠️ 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.

⚠️ 4 benchmarks measured no execution time

Nothing ran under measurement, usually because the compiler removed the code under test. These results are not comparable, so they count as unchanged.

Preventing compiler optimizations

⚡ 2 improved benchmarks
❌ 1 regressed benchmark
✅ 2229 untouched benchmarks
⏩ 329 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	take_fsl_random[128, 10]	32.7 µs	58.5 µs	-44.2%
⚡	Simulation	take_fsl_u8_random[256, 100]	98.6 µs	45.8 µs	×2.2
⚡	Simulation	take_fsl_nullable_random[16, 100]	97.1 µs	50.5 µs	+92.18%
⚠️	Simulation	take_fsl_f16_random[16, 100]	61.4 µs	< 1 ns	N/A
⚠️	Simulation	fixed_16_advancing_ptr_safe[100]	< 1 ns	< 1 ns	N/A
⚠️	Simulation	preverify_advancing_ptr_unchecked[1000]	< 1 ns	< 1 ns	N/A
⚠️	Simulation	preverify_advancing_ptr_unchecked[10000]	< 1 ns	< 1 ns	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 mk/bitpacked-stack-07-width-selection (d5ffca3) with mk/bitpacked-stack-06-v2-wire (60ce538)}

Footnotes

1. 329 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. ↩