Add vibe-aware playlists with CLAP text embeddings

Blend taste profile with text-embedded mood descriptions (e.g. "chill
ambient lo-fi") using pre-blended vector search against the existing
HNSW index. New optional `vibe` and `alpha` params on playlist generate
and recommendations endpoints. Backward compatible — no vibe = pure
taste profile (alpha=1.0).

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

src/haunt_fm/models/track.py

@@ -94,6 +94,8 @@ class Playlist(Base):
     name: Mapped[str | None] = mapped_column(Text)
     known_pct: Mapped[int] = mapped_column(Integer, nullable=False)
     total_tracks: Mapped[int] = mapped_column(Integer, nullable=False)
+    vibe: Mapped[str | None] = mapped_column(Text)
+    alpha: Mapped[float | None] = mapped_column(REAL)
     created_at: Mapped[datetime] = mapped_column(DateTime(timezone=True), server_default=func.now())
 
     tracks: Mapped[list["PlaylistTrack"]] = relationship(back_populates="playlist", cascade="all, delete-orphan")