Implement duplicate resolution strategy by quality and enhance related documentation

- Updated `duplicate_resolve.py` to introduce a new strategy for keeping files based on quality, considering resolution, source, codec, and size.
- Enhanced `planner.py` to utilize the new quality-based strategy during plan generation, updating quarantine reasons accordingly.
- Modified `README.md` to document the new `plan.duplicate_keep` options, including `by_quality`, and provided detailed descriptions of each strategy.
- Added unit tests in `test_duplicate_resolve.py` to validate the new quality-based resolution logic.
- Updated `analysis.json` and `plan.json` with new timestamps and IDs to reflect recent changes.

These updates improve the Video Library Manager's ability to handle duplicate files more effectively, ensuring users retain the highest quality versions.

CLAUDE.md

@@ -69,6 +69,7 @@ VLM follows a read-first, multi-stage pipeline:
 - `io.py` - Load/save identities JSON/CSV, plan/analysis input helpers
 - `utils.py` - UTC time, format_size, shared helpers
 - `analysis.py` - Completeness checking (episode gaps) and duplicate detection
+- `duplicate_resolve.py` - Duplicate group resolution (by_quality, by_reputation, first_seen, manual)
 - `planner.py` - Execution plan generation with conflict detection
 - `executor.py` - File operations (move/rename/quarantine) with rollback logging
 - `quarantine.py` - Quarantine management with manifest tracking
@@ -116,6 +117,7 @@ Key settings:
 - `log_level` - logging verbosity
 - `categories` - mapping of category names to directory name lists
 - `enrichment` (or `enrich`) - TMDB/api_keys, cache_db, translation, reputation; see README for full schema
+- `plan.duplicate_keep` - when using `vlm plan --analysis`: `by_quality` (resolution > source > codec > size), `by_reputation` (default), `first_seen`, or `manual`
 
 ### Category Mappings