Fix gap_vs_market computation in diag_orb_trending_features.py

Was using non-existent gap_pct key in enrichment dict; now computes from
today_open and prev_close fields. V30 results: hurst INVERTED+G2 FAIL,
ou_theta G1 FAIL (near-zero Pearson). gap_vs_market re-running as V30b.

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

apps/intraday_bt/scripts/diag_orb_trending_features.py

@@ -342,9 +342,13 @@ def main() -> None:
 
         # gap_vs_market: stock gap_pct - QQQ gap_pct on trade day
         enrich_day = enrichment.get(ticker, {}).get(date, {})
-        stock_gap = enrich_day.get("gap_pct")
+        s_open = enrich_day.get("today_open")
+        s_prev = enrich_day.get("prev_close")
+        stock_gap = ((s_open - s_prev) / s_prev) if (s_open and s_prev and s_prev > 0) else None
         qqq_day_enrich = qqq_enrich.get(date, {})
-        qqq_gap = qqq_day_enrich.get("gap_pct")
+        q_open = qqq_day_enrich.get("today_open")
+        q_prev = qqq_day_enrich.get("prev_close")
+        qqq_gap = ((q_open - q_prev) / q_prev) if (q_open and q_prev and q_prev > 0) else None
         f_gap_vs_mkt = (stock_gap - qqq_gap) if (stock_gap is not None and qqq_gap is not None) else None
 
         avg_vol = enrich_day.get("avg_daily_vol_14d")