feat(phase-4): LGBM 모델 + 앙상블 + 매칭/재학습 잡

- backend/app/models/lgbm.py: 종목 × horizon 별 LightGBM 회귀(y_ret_h)
  + 다중분류(y_dir_h, 3-class). joblib 으로 backend/data/models/{code}_h{H}_*.pkl
  저장. early_stopping(30). predict_one() 으로 최신 영업일 피처에 추론.
- backend/app/models/weights.py: ensemble_weights 테이블 IO,
  default w_chronos=0.6 / w_lgbm=0.4 (DB 행 없으면 fallback).
- backend/app/models/ensemble.py: Chronos sample 분포 + LGBM regression+cls
  결합. point/q10/q90 + prob_up/flat/down + direction 라벨. 한쪽 모델
  실패 시 다른 쪽 단독 fallback (cold start: chronos 단독).
- backend/app/pipelines/predict_one.py: predict_and_store(). 결과를
  predictions 테이블에 UPSERT, user_triggered 누적 OR. base_date = 마지막
  ohlcv 거래일, target_date = base_date + H 영업일(주말 스킵, 공휴일은
  매칭잡에서 자연 보정).
- backend/app/pipelines/match_outcomes.py: target_date == d 인
  user_triggered=TRUE 예측을 d 의 실제 종가와 매칭 → prediction_outcomes
  적재. direction_hit(±0.3% flat band) + abs_error. 실제 종가 없으면
  자연 skip.
- backend/app/pipelines/retrain_weekly.py: 시드 10종목 × H 재학습 +
  최근 30일 model_performance 적재.
- backend/app/db/migrations/003_ensemble_weights.sql: (code, horizon) →
  (w_chronos, w_lgbm, hit_rate_*, sample_count).
- backend/app/pipelines/scheduler.py:
    daily_batch    : 평일 16:00 KST
    match_outcomes : 평일 16:30 KST   ← 사용자가 확정한 매칭 시점
    retrain_weekly : 일요일 02:00 KST

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

backend/app/pipelines/scheduler.py

@@ -22,6 +22,8 @@ from apscheduler.triggers.cron import CronTrigger
 from pytz import timezone
 
 from app.pipelines.daily_batch import run_daily_batch
+from app.pipelines.match_outcomes import match_today
+from app.pipelines.retrain_weekly import run_weekly
 
 logger = logging.getLogger(__name__)
 KST = timezone("Asia/Seoul")
@@ -34,15 +36,34 @@ def start_scheduler() -> BackgroundScheduler:
     if _scheduler:
         return _scheduler
     _scheduler = BackgroundScheduler(timezone=KST)
+    # 16:00 평일: 시드 10종목 EOD/뉴스/공시/거시 갱신
     _scheduler.add_job(
         run_daily_batch,
-        CronTrigger(hour=16, minute=0, timezone=KST),
+        CronTrigger(day_of_week="mon-fri", hour=16, minute=0, timezone=KST),
         id="daily_batch_16",
         replace_existing=True,
         max_instances=1,
     )
+    # 16:30 평일: prediction_outcomes 매칭 배치
+    _scheduler.add_job(
+        match_today,
+        CronTrigger(day_of_week="mon-fri", hour=16, minute=30, timezone=KST),
+        id="match_outcomes_1630",
+        replace_existing=True,
+        max_instances=1,
+    )
+    # 일요일 02:00: LGBM 재학습 + 성능 기록
+    _scheduler.add_job(
+        run_weekly,
+        CronTrigger(day_of_week="sun", hour=2, minute=0, timezone=KST),
+        id="retrain_weekly_sun_0200",
+        replace_existing=True,
+        max_instances=1,
+    )
     _scheduler.start()
-    logger.info("scheduler started (daily_batch @ 16:00 KST)")
+    logger.info(
+        "scheduler started: daily_batch(16:00 mon-fri), match_outcomes(16:30 mon-fri), retrain_weekly(sun 02:00) KST"
+    )
     return _scheduler