Record authoritative macOS benchmark baseline

[tomsanbear]

Summary

• record the clean five-rotation M3 Pro macOS system-allocator baseline paired with the canonical Linux commit
• validate and publish separate, non-pooled Linux/macOS sections and platform-specific graphics
• extend deterministic report generation and CI for validated multi-platform inputs
• update the performance ledger and benchmark documentation

Validation

• validate_pair.py: comparable at 2dc82e0 with Rust 1.97.0
• reporting tests: 11 passed
• cargo check -p jack-vec-comparisons --all-targets
• cargo fmt --all -- --check
• git diff --check

Raw per-round measurements remain on the dedicated benchmark host; the compact authoritative JSON, Markdown, and deterministic SVG derivatives are versioned here.

[codspeed-hq]

Merging this PR will improve performance by 14.02%

⚡ 3 improved benchmarks
✅ 107 untouched benchmarks

Performance Changes

	Benchmark	BASE	HEAD	Efficiency
⚡	SmallVec4[4]	174.4 ns	145.3 ns	+20.08%
⚡	SmallVec8[8]	277.5 ns	248.3 ns	+11.74%
⚡	SmallVec4[8]	307.8 ns	278.6 ns	+10.47%

Tip

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

---

_{Comparing bench/canonical-macos-baseline (193cc06) with main (e95fa3e)}