Describes the restructured workflow where metadata gathering (TOC, CDDB,
MB, Vision-LLM) happens before ripping starts, so the user can review and
edit album.json before committing to the long rip — not after.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Concrete plan for Option B: new merger.py module with field-by-field
priority merging, duration_ms/disc_id model extensions, cover strategy,
and track-matching logic for sources with differing track counts.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Shorter, cleaner filenames consistent with Jellyfin conventions.
Updated all references in source, tests, and documentation.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
