feat(recommender): nightly ALS with offline metrics and CF candidates in the wave (M6)

- Implicit scores from play events and reactions per the signals table
  (90-day half-life, weaker skips outside the wave, dislikes override).
- numpy ALS for implicit feedback; time-split evaluation on the last
  7 days (NDCG@10, Recall@50) against a popularity baseline; a new model
  is activated only if NDCG@10 does not regress by more than 2%.
- Vectors are written to track_embeddings.cf and user_profiles_rec.taste_cf,
  versions to model_versions; NightlyTrainer runs at TRAIN_HOUR and on
  first start, `pipu-rec train` runs it manually.
- Wave: cf and item_cf candidate sources and a 0.10 cf term in the v1
  score; impressions are tagged heuristic-v1+cf.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

services/recommender/pipu_rec/settings.py

@@ -17,4 +17,5 @@ class Settings(BaseSettings):
     library_host_root: str = "/library-host"
     ffmpeg_path: str = "ffmpeg"
     contexts_path: str = Field(default="", description="contexts.yaml; пусто — встроенный")
+    train_hour: int = Field(default=3, ge=-1, le=23, description="час ночного обучения; -1 — выключено")
     version: str = Field(default="0.0.0-dev", alias="PIPU_VERSION")