perf: optimize cross-library move detection for single-library setups (#4888)

* feat: skip cross-library detection for single library setup

When only one library is configured, skip the cross-library move detection stage entirely as there are no other libraries to search in. This eliminates unnecessary database queries - the primary performance issue reported by users (5-6 hour scans with 13.5k missing files).

Implementation:
- Added library count check in processCrossLibraryMoves
- Returns input unchanged when len(state.libraries) == 1
- Logs debug message for troubleshooting

* refactor: use lightweight queries for cross-library move detection

Replace selectMediaFile() with newSelect() in FindRecentFilesByMBZTrackID and FindRecentFilesByProperties. These queries only need basic media file columns for hash and path comparisons, not annotations/bookmarks.

Benefits:
- Removes unnecessary LEFT JOINs with annotation and bookmark tables
- Reduces query overhead for cross-library file matching
- Follows existing pattern used by GetMissingAndMatching

The annotation/bookmark joins are user-specific (using loggedUser context) and unused in cross-library matching logic where only Equals() and IsEquivalent() checks are performed.

* test: add coverage for single-library and multi-library cross-library detection

Add test cases to verify:
1. Single-library setup correctly skips cross-library move detection
2. Multi-library setup continues to process cross-library moves

Implementation:
- New test verifies processCrossLibraryMoves returns input unchanged for single library
- Wrapped existing multi-library tests in Context with multiple libraries setup
- Ensures no regressions in multi-library matching behavior

Tests verify:
- Single-library: no database queries, input passed through unchanged
- Multi-library: cross-library matching still works correctly
- Reduces the likelihood of introducing single-library skip bugs in future

* fix: enhance cross-library detection by introducing totalLibraryCount

Signed-off-by: Deluan <deluan@navidrome.org>

---------

Signed-off-by: Deluan <deluan@navidrome.org>

scanner/scanner.go

@@ -28,11 +28,12 @@ type scannerImpl struct {
 
 // scanState holds the state of an in-progress scan, to be passed to the various phases
 type scanState struct {
-	progress        chan<- *ProgressInfo
-	fullScan        bool
-	changesDetected atomic.Bool
-	libraries       model.Libraries  // Store libraries list for consistency across phases
-	targets         map[int][]string // Optional: map[libraryID][]folderPaths for selective scans
+	progress          chan<- *ProgressInfo
+	fullScan          bool
+	changesDetected   atomic.Bool
+	libraries         model.Libraries  // Store libraries list for consistency across phases
+	targets           map[int][]string // Optional: map[libraryID][]folderPaths for selective scans
+	totalLibraryCount int              // Total number of libraries (unfiltered), for cross-library move detection
 }
 
 func (s *scanState) sendProgress(info *ProgressInfo) {
@@ -73,6 +74,7 @@ func (s *scannerImpl) scanFolders(ctx context.Context, fullScan bool, targets []
 		state.sendWarning(fmt.Sprintf("getting libraries: %s", err))
 		return
 	}
+	state.totalLibraryCount = len(allLibs)
 
 	if len(targets) > 0 {
 		// Selective scan: filter libraries and build targets map