refactor: make bitpacked chunk boundaries explicit

[mhk197]

Make constructors accept the offsets child explicitly, so slices, casts, patch reconstruction, and other array transformations preserve the stored boundaries. Slices retain their original offset origin and share the backing buffer; addresses are computed relative to the first boundary.

Introduce ChunkLayout, which borrows a materialized offsets buffer and derives each chunk's bit width from adjacent boundaries. It validates boundary ordering, 128-byte alignment of chunk sizes, supported widths, and the total packed span. Encoders may supply temporary widths to build the boundaries, but the persistent array contains only the offsets child.

The scalar bit_width remains available to existing kernels. Arrays must still have materialized, uniform offsets consistent with that scalar. This prepares the array API for a later migration of CPU readers; v1 serialization continues to omit the offsets child. CUDA changes only adapt the constructor API.

Validation: 395 FastLanes/BtrBlocks tests passed (1 skipped). Focused Clippy passed with all targets, all features, and warnings denied.

[codspeed]

Merging this PR will improve performance by 79.16%

⚠️ 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

⚡ 1 improved benchmark
✅ 2261 untouched benchmarks
⏩ 299 skipped benchmarks¹
🗄️ 1 archived benchmark run²

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
⚡	Simulation	take_fsl_random[128, 10]	58.5 µs	32.7 µs	+79.16%
⚠️	Simulation	take_fsl_u32_random[256, 10]	< 1 ns	< 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

Curious why performance improved? Comment @codspeedbot explain why performance improved on this PR, or directly use the CodSpeed MCP with your agent.

---

_{Comparing mk/bitpacked-stack-04-offset-child (06905e1) with mk/bitpacked-stack-03-width-child (9dc388a)}

Footnotes

1. 299 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. 1 benchmark was run, but is now archived. If it was deleted in another branch, consider rebasing to remove it from the report. Instead if it was added back, click here to restore it. ↩