feat(taste): track_tags schema + enrichment queries (Opt 1 foundation)

First step of taste-profile fidelity via metadata enrichment (milestone
#160, task #1490) — no ML sidecar, operator's constraint.

The taste profile's tag facet is built purely from raw ID3 tracks.genre
(splitGenres in internal/taste/profile.go). This lands the data layer for
enriching it with track-level folksonomy tags:

- track_tags(track_id, tag, weight) — a global cache of style/mood tags,
  top-K per track, weight = normalized folksonomy strength [0,1].
- tracks.tag_source / tag_sources_version — versioned enrichment
  bookkeeping mirroring artists.artist_art_source (NULL = eligible,
  provider name = found, 'none' = settled, version bump = re-process).
- Queries: ListTracksMissingTags (batch drainer), DeleteTrackTags +
  InsertTrackTag (atomic per-track replace), SetTrackTagSource, and
  ListPlayed/LikedTrackTagsForUser for the recompute to union enriched
  tags into the tag facet alongside genre.

No consumer yet — the enricher (MusicBrainz + Last.fm providers,
track-level) and the taste-recompute integration land next.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

internal/db/queries/track_tags.sql

@@ -0,0 +1,57 @@
+-- Track-tag enrichment queries (milestone #160, #1490). The tag enricher
+-- drains ListTracksMissingTags, fetches from the provider chain, and
+-- writes the merged top-K via DeleteTrackTags + InsertTrackTag, then
+-- stamps SetTrackTagSource. Mirrors the coverart artist-enricher shape.
+
+-- name: ListTracksMissingTags :many
+-- Tracks eligible for tag enrichment: never processed (tag_source NULL)
+-- or previously settled 'none' under an older provider version. Returns
+-- the fields the provider chain needs — recording MBID (nullable) for
+-- keyed lookups, plus title + artist name for name-based fallback.
+-- $1 = current tag_sources_version, $2 = limit.
+SELECT t.id, t.mbid, t.title, a.name AS artist_name
+  FROM tracks t
+  JOIN artists a ON a.id = t.artist_id
+ WHERE t.tag_source IS NULL
+    OR (t.tag_source = 'none' AND t.tag_sources_version < $1)
+ ORDER BY t.id
+ LIMIT $2;
+
+-- name: DeleteTrackTags :exec
+-- Clear a track's cached tags before rewriting (atomic replace by the caller).
+DELETE FROM track_tags WHERE track_id = $1;
+
+-- name: InsertTrackTag :exec
+-- Upsert one (track, tag); keep the stronger weight when two providers
+-- agree on a tag with different folksonomy strengths.
+INSERT INTO track_tags (track_id, tag, weight)
+VALUES ($1, $2, $3)
+ON CONFLICT (track_id, tag)
+    DO UPDATE SET weight = GREATEST(track_tags.weight, EXCLUDED.weight);
+
+-- name: SetTrackTagSource :exec
+-- Stamp the enrichment outcome so the batch drainer skips settled rows.
+-- $2 = 'lastfm' | 'musicbrainz' | 'mixed' | 'none', $3 = current version.
+UPDATE tracks SET tag_source = $2, tag_sources_version = $3 WHERE id = $1;
+
+-- name: ListPlayedTrackTagsForUser :many
+-- The enriched tags for every track this user played inside the taste
+-- window, so the recompute can union them into the tag facet alongside
+-- ID3 genre. $1 = user_id, $2 = window_days. One row per (track, tag).
+SELECT tt.track_id, tt.tag, tt.weight
+  FROM track_tags tt
+ WHERE tt.track_id IN (
+       SELECT DISTINCT pe.track_id
+         FROM play_events pe
+        WHERE pe.user_id = $1
+          AND pe.was_skipped = false
+          AND pe.started_at > now() - make_interval(days => $2::int)
+   );
+
+-- name: ListLikedTrackTagsForUser :many
+-- Enriched tags for the user's liked tracks — the like-bonus path in the
+-- recompute mirrors the play path, so it needs the same tag lookup.
+SELECT tt.track_id, tt.tag, tt.weight
+  FROM track_tags tt
+  JOIN general_likes gl ON gl.track_id = tt.track_id
+ WHERE gl.user_id = $1;