Fix parity check: always compare lead-in rate + auto-resolve past alerts

- DISTINCT ON tiebreaker was undefined when scraper writes multiple room
  types in one batch (same scraped_at); adding rate_gross ASC ensures we
  always pick the cheapest (lead-in / best available) rate, matching like
  for like against the Newbook BAR tariff
- Past-date active alerts were never touched (start = today meant they
  fell outside the query window); now resolved at the top of each run

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

backend/jobs/check_rate_parity.py

@@ -175,7 +175,7 @@ def gather_comparisons(db, start: date, end: date, cfg: dict) -> list:
         WHERE h.tier = 'own'
           AND r.rate_date BETWEEN :fd AND :td
           AND r.rate_gross IS NOT NULL
-        ORDER BY r.rate_date, r.scraped_at DESC
+        ORDER BY r.rate_date, r.scraped_at DESC, r.rate_gross ASC
     """), {"fd": start, "td": end}).mappings().all()
 
     newbook_rows = db.execute(text("""
@@ -252,8 +252,17 @@ def run_parity_check() -> dict:
             logger.info("Parity check skipped (disabled)")
             return {"status": "disabled"}
 
-        start = date.today()
-        end = start + timedelta(days=HORIZON_DAYS)
+        today = date.today()
+        start = today
+        end = today + timedelta(days=HORIZON_DAYS)
+
+        # Resolve any lingering active alerts for dates that have already passed
+        db.execute(text("""
+            UPDATE rate_parity_alerts
+            SET alert_status = 'resolved'
+            WHERE alert_status = 'active'
+              AND rate_date < :today
+        """), {"today": today})
 
         comparisons = gather_comparisons(db, start, end, cfg)