perf(cuda): pipeline local file reads through cacheable pinned buffers

[0ax1]

Pipeline large local CUDA reads in 4 MiB chunks, transferring completed chunks into one device allocation without waiting for the full read.

• Use cacheable pooled pinned memory for staging.
• Bound concurrent host reads per file and retain buffers safely through cancellation and asynchronous transfers.
• Preserve requested byte ranges, excluding direct-I/O alignment padding from chunked transfers.
• Add regression coverage for chunk placement, short reads, overflow, cancellation, and buffer reuse.

[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.

⚠️ 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

⚡ 3 improved benchmarks
❌ 1 regressed benchmark
✅ 2243 untouched benchmarks
⏩ 176 skipped benchmarks¹
🗄️ 1 archived benchmark run²

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	WallTime	arrow_checked_add_u32_neon[16384]	12.9 µs	20.3 µs	-36.5%
⚡	Simulation	decode_primitives[f32, (1000, 512)]	64.7 µs	42.5 µs	+52.26%
⚡	Simulation	decompress[u64, (4000, 1024)]	87.1 µs	71.6 µs	+21.8%
⚡	Simulation	allocate_drop_bytes[0]	635.5 ns	527.2 ns	+20.55%

Tip

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

---

_{Comparing ad/cuda-pipelined-local-reads (314f2df) with develop (75b6225)}

Footnotes

1. 176 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. ↩