perf(multitude): add arc_array allocation/timing benchmark

[martintmk]

Add criterion_arc_array bench comparing construction of an Arc<[Arc<[u8]>]> (8 x 16B properties) via the global allocator vs the multitude arena, using the alloc_tracker crate for per-iteration allocation volume. Two strategies per path: build through a growable vec (*) and build directly from a pre-created slice (*_from_slice).

The bench surfaces a pathology in the arena vec-then-freeze path: under a warmed, reset-and-reused arena it re-grabs a full 64 KiB chunk every iteration and runs ~35x slower than the global allocator. The direct alloc_slice_clone_arc path reuses chunks correctly (0 bytes/op).

variant	time	bytes/op	allocs/op
global	~562 ns	528	10
arena	~20.0 µs	65535	1
global_from_slice	~143 ns	144	1
arena_from_slice	~146 ns	0	0

[martintmk]

this is ugly, we should have into_arc here

[Copilot]

⚠️ Not ready to approve

The new benchmark’s module-level documentation misdescribes what the arena “from_slice” path actually allocates, which can mislead interpretation of the results.

Pull request overview

Adds a new Criterion benchmark to the multitude crate to compare building an Arc<[Arc<[u8]>]> via the global allocator vs the multitude arena, including per-iteration allocation tracking with alloc_tracker, to highlight a warm-arena reuse pathology in the vec-then-freeze path.

Changes:

• Add alloc_tracker as a dev-dependency for crates/multitude and register a new criterion_arc_array benchmark target.
• Introduce criterion_arc_array bench that times two construction strategies (vec-then-freeze vs from-slice) and prints allocation volume via alloc_tracker.
• Update Cargo.lock to include the new dev dependency.

File summaries

File	Description
crates/multitude/Cargo.toml	Adds alloc_tracker dev-dep and registers the new Criterion bench.
crates/multitude/benches/criterion_arc_array.rs	New benchmark measuring time + allocation volume for global vs arena construction paths.
Cargo.lock	Locks alloc_tracker as a dependency of multitude (dev).

Copilot's findings

• Files reviewed: 2/3 changed files
• Comments generated: 1

Note

Your feedback helps us improve the quality of this feature.
Please use 👍 or 👎 to tell us whether this assessment is correct.

---

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

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.0%. Comparing base (b190aba) to head (9cc57ca).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #507   +/-   ##
=======================================
  Coverage   100.0%   100.0%           
=======================================
  Files         343      343           
  Lines       26121    26121           
=======================================
  Hits        26121    26121           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.