backtest: explain the regime-gated empty window + toggle to view raw-criteria trades

Owner saw a blank /backtest — it was CORRECT (the regime gate sat out every
recent session) but unexplained. Now: ?regime=0|1 override on /api/backtest,
an 'apply regime gate (as deployed)' checkbox, and an empty-state message
saying WHY there are no trades and what to click. Gated view = what the bot
would do; ungated = the raw simple criteria.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

tests/unit/api/test_backtest_endpoint.py

@@ -244,3 +244,24 @@ def test_replay_window_replays_live_track_only():
     from research.experiments.replay import replay_window
     out = replay_window(store, str(SESS1), str(SESS2))
     assert len(out["trades"]) == 4  # the 4 v1 fires only; liq1 excluded
+
+
+def test_backtest_regime_override_bypasses_gate(monkeypatch):
+    """/backtest UX fix (owner: 'no data in backtest'): when the regime gate
+    zeroes a whole window the page must be able to show RAW-criteria trades
+    via ?regime=0, and the response must echo which mode ran."""
+    import api.main as m
+    store = _seed_store()
+    monkeypatch.setenv("TRADING_REGIME_GATE_ENABLED", "1")
+    # unfavorable scores for BOTH sessions -> gated run has zero trades
+    from research import regime as rg
+    rg.apply_schema(store.con)
+    for sess in (SESS1, SESS2):
+        store.con.execute(
+            "INSERT INTO regime_scores (model_version, session_date, p_profit, favorable, computed_at) "
+            "VALUES ('meta_v1', ?, 0.1, false, now()) ON CONFLICT DO NOTHING", [sess])
+    state = m.DashboardState("momentum")
+    gated = state.backtest(days=10)
+    raw = state.backtest(days=10, regime_override=False)
+    assert gated["summary"]["n"] == 0 and gated["settings"]["regime_gate"] is True
+    assert raw["summary"]["n"] == 4 and raw["settings"]["regime_gate"] is False