test(pm): cover resolver demand mainloop

[elrrrrrrr]

Summary

• add focused unit coverage for demand/preload queue priority and promotion
• cover manifest state cache updates, waiter wakeups, speculative full-manifest extracts, and transitive prefetch enqueueing
• add a regression test for non-semver exact-version extraction single-flight behavior

Validation

• cargo fmt
• cargo check -p utoo-ruborist
• cargo test -p utoo-ruborist resolver::builder::tests::test_fetch_queues -- --nocapture
• cargo test -p utoo-ruborist resolver::builder::tests::test_apply_fetch_result -- --nocapture
• cargo test -p utoo-ruborist resolver::builder::tests::test_non_semver_exact_version_extract_single_flight -- --nocapture
• cargo test -p utoo-ruborist resolver::builder::tests::test_enqueue_version -- --nocapture
• cargo test -p utoo-ruborist resolver::builder::tests::test_schedule_registry_fetch_dedupes_semver_request -- --nocapture
• cargo clippy --all-targets -- -D warnings --no-deps

Split Plan

This is split out of #3043 so the behavior switch PR stays focused on implementation while the detailed demand-mainloop coverage is reviewed separately.

[gemini-code-assist]

Code Review

This pull request introduces a comprehensive suite of unit tests for the resolver's manifest fetching and caching logic in crates/ruborist/src/resolver/builder.rs. These tests cover job deduping, priority handling (Demand vs. Prefetch), concurrency limits, and caching behavior for both exact versions and full manifests. A CountingRegistry mock was also added to facilitate tracking of manifest jobs during testing. I have no feedback to provide.

[github-actions]

📊 pm-bench-phases · f76f055 · linux (ubuntu-latest)

Workflow run — ant-design

PMs: utoo (this branch) · utoo-npm (latest published) · bun (latest)

npmjs.org

p0_full_cold

PM	wall	±σ	user	sys	RSS	pgMinor
bun	7.54s	0.93s	10.31s	6.15s	669M	298.8K
utoo-next	7.35s	1.54s	10.59s	7.85s	984M	130.4K
utoo-npm	7.51s	1.45s	10.84s	8.07s	1.01G	119.7K
utoo	6.78s	0.13s	11.47s	7.90s	959M	146.0K

PM	vCtx	iCtx	netRX	netTX	cache	node_mod	lock
bun	17.2K	17.6K	1.19G	6M	1.86G	1.75G	1M
utoo-next	138.1K	93.7K	1.16G	5M	1.71G	1.70G	2M
utoo-npm	152.7K	106.8K	1.16G	5M	1.71G	1.70G	2M
utoo	125.4K	67.7K	1.16G	6M	1.71G	1.70G	2M

p1_resolve

PM	wall	±σ	user	sys	RSS	pgMinor
bun	2.02s	0.03s	4.43s	0.77s	527M	165.6K
utoo-next	2.85s	0.15s	5.35s	1.20s	612M	87.7K
utoo-npm	2.83s	0.04s	5.54s	1.44s	617M	78.6K
utoo	2.24s	0.07s	6.11s	1.15s	651M	123.7K

PM	vCtx	iCtx	netRX	netTX	cache	node_mod	lock
bun	9.8K	4.8K	203M	3M	107M	-	1M
utoo-next	55.3K	71.8K	199M	2M	7M	3M	2M
utoo-npm	80.6K	92.8K	200M	2M	7M	3M	2M
utoo	16.0K	17.9K	202M	3M	7M	3M	2M

p3_cold_install

PM	wall	±σ	user	sys	RSS	pgMinor
bun	5.34s	0.13s	5.89s	6.08s	592M	197.4K
utoo-next	6.29s	2.06s	4.91s	6.91s	540M	60.5K
utoo-npm	6.77s	2.52s	4.98s	6.92s	529M	64.1K
utoo	4.82s	0.40s	4.91s	6.78s	474M	62.2K

PM	vCtx	iCtx	netRX	netTX	cache	node_mod	lock
bun	6.3K	6.2K	1019M	4M	1.76G	1.76G	1M
utoo-next	126.1K	53.5K	990M	3M	1.70G	1.70G	2M
utoo-npm	131.4K	47.2K	990M	4M	1.70G	1.70G	2M
utoo	119.7K	49.8K	989M	3M	1.70G	1.70G	2M

p4_warm_link

PM	wall	±σ	user	sys	RSS	pgMinor
bun	2.11s	0.05s	0.14s	1.17s	134M	32.8K
utoo-next	1.84s	0.03s	0.48s	2.33s	78M	18.4K
utoo-npm	1.80s	0.09s	0.48s	2.35s	79M	18.4K
utoo	1.90s	0.25s	0.48s	2.32s	79M	18.7K

PM	vCtx	iCtx	netRX	netTX	cache	node_mod	lock
bun	238	15	5M	19K	1.91G	1.75G	1M
utoo-next	42.1K	18.6K	7K	11K	1.70G	1.70G	2M
utoo-npm	42.7K	18.3K	4K	9K	1.70G	1.70G	2M
utoo	42.7K	17.9K	2K	5K	1.71G	1.70G	2M

npmmirror.com: no output captured.