chore(bench): skip multi-threaded cases in CodSpeed memory mode

[stormslowly]

Why

CodSpeed memory mode runs Valgrind massif, which conflates allocations across worker threads and produces noisy, non-actionable deltas for the multi-threaded resolver benches. Skip them in that mode so memory deltas only come from the single-threaded cases.

What

1. Rename the two threaded bench cases to share a [multi-threaded] prefix (matches the existing [single-threaded]resolve with many extensions style):
   • multi-thread → [multi-threaded]resolve
   • resolve from symlinks multi thread → [multi-threaded]resolve from symlinks
2. Gate them on the CODSPEED_RUNNER_MODE env var (set by the CodSpeed action). They keep running in the simulation step and are skipped only in memory mode.

No workflow change needed — the existing mode: memory action already exports CODSPEED_RUNNER_MODE=memory into the bench process.

[Copilot]

Pull request overview

This PR standardizes the naming of the multi-threaded Criterion benchmark cases in benches/resolver.rs by adding a shared [multi-threaded] prefix, enabling easy selection/skipping via regex (e.g., when running CodSpeed in modes where multi-threaded throughput isn’t meaningful).

Changes:

• Renamed the "multi-thread" benchmark case to "[multi-threaded]resolve".
• Renamed "resolve from symlinks multi thread" to "[multi-threaded]resolve from symlinks".

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[codspeed-hq]

Merging this PR will improve performance by 3.2%

⚡ 1 improved benchmark
✅ 87 untouched benchmarks
🆕 4 new benchmarks
⏩ 4 skipped benchmarks¹

Performance Changes

	Mode	Benchmark	BASE	HEAD	Efficiency
🆕	Memory	resolver[[multi-threaded]resolve from symlinks]	N/A	11.6 MB	N/A
🆕	Memory	resolver[[multi-threaded]resolve]	N/A	10.9 MB	N/A
🆕	Simulation	resolver[[multi-threaded]resolve]	N/A	59 ms	N/A
🆕	Simulation	resolver[[multi-threaded]resolve from symlinks]	N/A	76.5 ms	N/A
⚡	Simulation	specifier/branches[multi-question]	940 ns	910.8 ns	+3.2%

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 chore/rename-multi-threaded-bench (deeba50) with main (4bdda7b)}

Footnotes

1. 4 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. ↩