perf(library): send the track list as a dictionary-encoded column table

GetTracks answered with one JSON object per track: 20.5 MB at 26k
tracks, ~350 bytes a row of repeated key names, four cover URLs that
are identical across an album, and the album, artist and genre strings
repeated on every track. Encoding it was ~170 MB of transient Go
allocation, and parsing it the WebView's memory peak.

GetTrackTable replaces it: one array per column, every repeated
string stored once and sent as an index, genre lists interned, and only
the fields the Tracks view reads. LastPlayed and the three larger cover
tiers are left to the details dialog, which now fetches whole tracks by
path. Each row still goes through trackFromRow, so this is an encoding
of the one projection, not a second one. 167 bytes a track against 929
in the test library; a Go test holds it under a quarter of the object
encoding, and a Vitest test reads the Go columns and the generated
Track interface and fails if they drift.

An empty library is now an empty table rather than the "no tracks in
library" error the old binding returned.

Closes #281

backend/library/scan_fixtures_test.go

@@ -58,15 +58,15 @@ func TestScan_FixtureLibraryLeavesNothingBehind(t *testing.T) {
 		t.Skip("fixture library is empty; run make testdata")
 	}
 
-	tracks, err := lib.GetTracks(0)
+	table, err := lib.GetTrackTable(0)
 	if err != nil {
-		t.Fatalf("GetTracks: %v", err)
+		t.Fatalf("GetTrackTable: %v", err)
 	}
 
 	// One track per file: the projection cannot multiply rows, because
 	// there is no join table left to multiply them.
-	if int64(len(tracks)) != files {
-		t.Errorf("GetTracks returned %d rows for %d files", len(tracks), files)
+	if int64(len(table.FilePath)) != files {
+		t.Errorf("GetTrackTable returned %d rows for %d files", len(table.FilePath), files)
 	}
 
 	// Nothing shared outlives what refers to it.