Feat: Add performance stress tests and documentation polish (P5)

[grunch]

Add performance/stress tests with 50 custom nodes covering add/persist, selectedNode lookup, mid-list removal, batch metadata update, and init timing. Update CLAUDE.md with multi-Mostro architecture section. Expand Phase 5 documentation in MULTI_MOSTRO_SUPPORT.md with implementation details and updated test coverage (74 total).

Summary by CodeRabbit

• New Features

  • Completed multi-Mostro instance support, enabling seamless connection to and switching between multiple nodes with persistent configuration.

• Documentation

  • Updated documentation marking the multi-node feature as completed with comprehensive implementation details and workflow specifications.

• Tests

  • Added performance test suite (5 tests) validating stability and efficiency when managing large-scale node lists.

[coderabbitai]

Walkthrough

This PR documents the completion of Phase 5 multi-Mostro instance support feature, adds comprehensive architectural documentation in CLAUDE.md describing node connection and management workflows, and introduces a new performance test suite with five stress tests validating large-scale node operations and persistence behavior.

Changes

Cohort / File(s)	Summary
Documentation CLAUDE.md, docs/MULTI_MOSTRO_SUPPORT.md	Added Multi-Mostro Instance Support architectural documentation covering trusted/custom node types, metadata handling, and storage strategies. Updated Phase 5 status to "Completed" with expanded goals including performance stress tests and detailed work item tracking with implementation notes.
Testing test/features/mostro/mostro_nodes_performance_test.dart	New performance test suite for MostroNodesNotifier with five stress tests: large node list persistence (50 nodes), node selection and persistence, node removal integrity, batch metadata updates across node types, and initialization performance benchmarking under 1 second.

Estimated code review effort

🎯 3 (Moderate) | ⏱️ ~18 minutes

Possibly related PRs

• MostroP2P/mobile\Feat: Add multi-Mostro node data model and trusted nodes registry (P1) #436: Introduces the multi-Mostro data model and notifier that these tests directly exercise and validate.
• MostroP2P/mobile\Feat: Add multi-Mostro integration tests (P4) #443: Modifies MostroNodesNotifier behaviors (add/select/remove) that the new performance tests directly validate at scale.
• MostroP2P/mobile\Feat: Add kind 0 metadata fetching for Mostro nodes (P2) #437: Updates metadata handling in MostroNode that the batch metadata update tests verify across node types.

Suggested reviewers

• AndreaDiazCorreia
• mostronator
• BraCR10

Poem

🐰 Whiskers twitch with delight, performance tests hop in sight,
Multi-node paths now documented so bright,
Phase Five complete, the feature's just right,
Fifty nodes dance through persistence's flight,
A rabbit's delight! 🌟

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Description Check	✅ Passed	Check skipped - CodeRabbit’s high-level summary is enabled.
Title check	✅ Passed	The title accurately summarizes the main changes: performance stress tests and documentation updates for Phase 5.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment
• Commit unit tests in branch phase-5-polish-edge-cases

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[coderabbitai]

Caution

Some comments are outside the diff and can’t be posted inline due to platform limitations.

⚠️ Outside diff range comments (1)

CLAUDE.md (1)

583-583: ⚠️ Potential issue | 🟡 Minor

Stale "Last Updated" date.

This line still reads "October 8, 2025" but the file is being modified now (February 2026). Consider updating to the current date.

Proposed fix

-**Last Updated**: October 8, 2025
+**Last Updated**: February 11, 2026

🧹 Nitpick comments (2)

test/features/mostro/mostro_nodes_performance_test.dart (2)

158-192: Iterating notifier.state while mutating state — safe but subtle.

The for (final node in notifier.state) loop captures the list reference at iteration start. Since updateNodeMetadata replaces state with a new list, the loop iterates the stale snapshot — which is fine here because you only read node.pubkey. However, this subtlety could confuse future maintainers.

Consider capturing the list explicitly for clarity:

Optional: make the snapshot explicit

-      for (final node in notifier.state) {
+      final snapshot = List.of(notifier.state);
+      for (final node in snapshot) {

---

194-224: Timing assertion tests mocked I/O, not real SharedPreferences.

The < 1000ms threshold is a reasonable regression guard against algorithmic blowup, but since MockSharedPreferencesAsync is backed by an in-memory Map, it doesn't measure actual disk I/O latency. The test name and the docs (line 208 in MULTI_MOSTRO_SUPPORT.md: "Init with 50 pre-existing nodes completes promptly") could be slightly misleading. Worth noting in a comment that this validates algorithmic/parsing performance, not storage I/O.

[mostronator]

Clean Phase 5 completion! 👏

Performance tests:

• 5 stress tests with 50 nodes covering add/persist, lookup, mid-list removal, batch metadata, and init timing
• makePubkey() helper is a nice pattern for deterministic test data
• Init timing assertion (<1s) is a good safeguard against O(n²) regressions
• Persistence round-trip verification (create notifier1 → populate → create notifier2 → verify) is thorough

CLAUDE.md:

• Multi-Mostro architecture section is well-structured — covers all the key design decisions in bullet form

Docs:

• Phase 5 expanded with clear work item analysis table showing what was done where
• Test coverage summary updated (74 total)

LGTM! 🚀

[AndreaDiazCorreia]

LGTM, clean perf tests with good persistence round-trip coverage, docs look solid, ACK