bench(pco): scalar access baseline for CodSpeed

[joseph-isaacs]

One divan benchmark, scalar_access, reading scalars out of a PCO array through execute_scalar, to give CodSpeed a baseline before the scalar-probe work lands. The probe PR will switch the repeated cases to repeated_probe() under the same bench names so the comparison shows up on merge.

Fixture: 16 384 u32 rows with values row / 16, PCO level 8, 1024 values per page. Nullable cases null every 11th row. Indices come from a fixed-seed LCG, clustered within 256 rows starting at row 4096 (one page) or scattered over the whole array.

Cases are (access_count, nullable, scattered):

Case	Local median
(1, false, false)	6.2 µs
(1024, false, false)	5.70 ms
(1024, true, false)	4.25 ms
(1024, false, true)	5.61 ms

Today every read decodes its page, so clustered and scattered cost the same. The single-read case is the control for the one-off path, which the probe work does not change.

Overlaps with #9895, which covers the same ground with cross-chunk and in-chunk cases; only one of the two should merge.

Checks

• cargo bench -p vortex-pco --bench scalar: builds and runs
• clippy and fmt not run locally

🤖 Generated with Claude Code

[codspeed]

Merging this PR will regress 2 benchmarks

⚠️ 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

⚡ 4 improved benchmarks
❌ 2 regressed benchmarks
✅ 2192 untouched benchmarks
🆕 4 new benchmarks
⏩ 218 skipped benchmarks¹

Warning

Please fix the performance issues or acknowledge them on CodSpeed.

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
❌	Simulation	random_i8[0.5]	71.3 µs	94.9 µs	-24.81%
❌	Simulation	new_raw_prim_test_between[i32, 2048]	64.1 µs	77.9 µs	-17.66%
⚡	Simulation	execute_scalar_struct_wide	503 µs	389.6 µs	+29.12%
⚡	Simulation	random_i16[0.8]	96.4 µs	78.5 µs	+22.83%
⚡	Simulation	execute_scalar_struct_simple	106.5 µs	89.4 µs	+19.02%
⚡	Simulation	allocate_drop_arrow[0]	456.9 ns	402.7 ns	+13.45%
🆕	Simulation	scalar_access[(1, false, false)]	N/A	208.8 µs	N/A
🆕	Simulation	scalar_access[(1024, false, false)]	N/A	132.3 ms	N/A
🆕	Simulation	scalar_access[(1024, false, true)]	N/A	133.3 ms	N/A
🆕	Simulation	scalar_access[(1024, true, false)]	N/A	100.7 ms	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 ji/pco-scalar-bench (7403cf4) with develop (b5f43ba)²}

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. ↩

2. No successful run was found on develop (d69291f) during the generation of this report, so b5f43ba was used instead as the comparison base. There might be some changes unrelated to this pull request in this report. ↩