fix: normalize AuDD scores + add filter to admin UI (#413)
* fix: normalize AuDD scores + add filter to admin UI
- normalize scores from integer (0-100) to float (0.0-1.0) when storing
context, fixing the 0% display issue for potential matches
- add filter controls to admin UI: pending (default), resolved, all
- preserve current filter when resolving flags or refreshing
- add unit test for score normalization
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
* fix: show match count instead of misleading 0% scores
AuDD doesn't return confidence scores when using accurate_offsets mode
(which we use for timecode data). All scores were 0, showing as "0% match".
Changed to show "{N} matches" badge instead, which is accurate and useful.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Co-Authored-By: Claude <noreply@anthropic.com>
---------
Co-authored-by: Claude <noreply@anthropic.com>
authored by
zzstoatzz.io
Claude
and committed by
GitHub
32711970
99352dce