Add defensible orphaned audio detection

serato_doctor/models/health.py

@@ -18,6 +18,7 @@ class HealthReport:
     suspected_cloud_conflict_groups: int
     suspected_cloud_conflict_files: int
     unused_tracks: int
+    orphan_candidates: Optional[int]
     suggested_matches: int
     broken_symlinks: int
     static_crates: int