feat(mind): inline suspicious transcript segment highlights

[ucguy4u]

Summary

• Suspicious transcript segments (from TranscriptQualityEvaluator) now show amber background + border directly in the transcript list
• Warning icon (⚠️) on each flagged segment opens a dialog with the specific quality signals (e.g. "low-confidence word shapes", "abnormally long words", "repeated tokens")
• Evidence highlights retain priority — if a segment is both suspicious and highlighted for evidence, the evidence style wins
• Zero extra data fetching: reads from MeetingTranscriptQualityReport.segmentIssues already loaded for the review banner

Changes

• MeetingIrTranscriptList — new suspiciousSegmentIds + suspiciousSignals params; extracted _buildSegmentRow + _showQualitySignals dialog
• MeetingScreen — two computed getters feeding into the list widget

Test plan

• app/tool/run_mind_macos.sh full rebuild
• Import a voice memo with poor audio quality; verify amber-highlighted segments appear
• Tap the ⚠️ icon; verify dialog shows the correct signals
• Tap a segment with evidence (from MoM) that is also suspicious; verify tertiary (evidence) color wins
• High-quality transcript (all segments pass evaluator): verify no amber rows

🤖 Generated with Claude Code

[github-actions]

Plugin Module Size Gate

Policy: modules over 3 MB must be delivered as plugins; plugin modules over 5 MB must document cache management.

Module	Size	Type	Status
packages/feature_mind	2.94 MB	bundled	WARN: 1-3 MB requires PR size justification

[github-actions]

🚀 PR Quick Check Summary

Check	Status	Description
PR Validation	❌ failure	Title format, docs, bundled model guardrail
Code Quality	❌ failure	Analyze, formatting
Core Tests	❌ failure	Core package unit tests
Airo Mind Shell	❌ failure	Mind flavour analyze, tests, entrypoint compile

💡 Note: Full app tests, coverage reports, and security scans run on merge to main.

View Details