phase 7-8 完成 + warmup 修复 + 产物结构化重组

主要内容：
- Phase 8 PROMOTE: finalist #1 (trial #324) registry 条目，自动生成
- Optuna objective warmup bug 修复 (shared/optimizer/objective.py)
- studies/ 目录按用途重组为 optuna/ + finalists/ + features/ 三层
- reports/ 加入 Optuna 中文 dashboard (5 主图 + 18 slice + 15 contour)
- 新增 PROJECT_GUIDE.md 项目说明文档
- 新增 build_registry_entry.py / build_optuna_dashboard.py / build_feature_datasets.py
- .gitignore: 允许提交 studies/*.db (Optuna DB) 和 reports/*.html (MT5 + dashboard)

scripts/diag_signal_trigger.py

@@ -0,0 +1,72 @@
+"""Check indicator values + signal conditions at 2025-01-02 01:50 vs 15:45.
+
+MT5 first trade fired 2025.01.02 01:55 LONG @ 2624.05
+  → signal bar 01:50 close=2623.84, fill at 01:55 open=2623.84
+Python first trade fired 2025-01-02 15:50 LONG @ 2642.57
+  → signal bar 15:45 close=2642.54, fill at 15:50 open=2642.54
+
+Both engines should fire same signals on same bars. Why do they differ?
+"""
+import sys
+from pathlib import Path
+PROJECT = Path(__file__).resolve().parent.parent
+sys.path.insert(0, str(PROJECT))
+
+import numpy as np
+import pandas as pd
+
+from shared.indicators.base import atr, ema, rsi
+from shared.data.loaders import load_bars
+from strategies.gold_scalper_pro.search_space import FROZEN_BASELINE
+import optuna
+
+# Load finalist #1 params
+db = PROJECT / "studies" / "optuna" / "gold_scalper_pro_is2025.db"
+study = optuna.load_study(
+    study_name="gold_scalper_pro_is2025",
+    storage=f"sqlite:///{db}",
+)
+from shared.optimizer.selector import select_diverse_topn
+from strategies.gold_scalper_pro.search_space import SEARCH_SPACE
+finalists = select_diverse_topn(study, n=3, ranges=SEARCH_SPACE)
+params = {**FROZEN_BASELINE, **finalists[0].params}
+
+print(f"InpFastEmaPeriod={params['InpFastEmaPeriod']}, "
+      f"InpSlowEmaPeriod={params['InpSlowEmaPeriod']}, "
+      f"InpRsiPeriod={params['InpRsiPeriod']}, "
+      f"InpAtrPeriod={params['InpAtrPeriod']}")
+print(f"InpRsiBuyLevel={params['InpRsiBuyLevel']}, "
+      f"InpPullbackAtrMult={params['InpPullbackAtrMult']}")
+
+m5 = load_bars(PROJECT / "data" / "XAUUSD_M5_2024-06-26_2026-06-26.parquet")
+window = m5[(m5["timestamp"] >= "2024-12-01 00:00:00") & (m5["timestamp"] < "2025-01-03 00:00:00")].reset_index(drop=True)
+
+close = window["close"].to_numpy(dtype=float)
+high = window["high"].to_numpy(dtype=float)
+low = window["low"].to_numpy(dtype=float)
+
+fast = ema(close, int(params["InpFastEmaPeriod"]))
+slow = ema(close, int(params["InpSlowEmaPeriod"]))
+rsi_arr = rsi(close, int(params["InpRsiPeriod"]))
+atr_arr = atr(high, low, close, int(params["InpAtrPeriod"]))
+
+# Find rows on 2025-01-02 around 01:50 and 15:45
+window["fast"] = fast
+window["slow"] = slow
+window["rsi"] = rsi_arr
+window["atr"] = atr_arr
+window["trend_up"] = (fast > slow) & (close > slow)
+window["near_fast"] = np.abs(close - fast) <= (params["InpPullbackAtrMult"] * atr_arr)
+window["rsi_prev"] = np.roll(rsi_arr, 1)
+window["buy_cross"] = (window["rsi_prev"] < params["InpRsiBuyLevel"]) & (rsi_arr >= params["InpRsiBuyLevel"])
+window["buy_signal"] = window["trend_up"] & window["near_fast"] & window["buy_cross"]
+
+print("\n=== Around 2025-01-02 01:45-02:00 ===")
+sub = window[(window["timestamp"] >= "2025-01-02 01:45:00") & (window["timestamp"] <= "2025-01-02 02:00:00")]
+print(sub[["timestamp", "open", "high", "low", "close", "fast", "slow",
+          "rsi", "atr", "trend_up", "near_fast", "buy_cross", "buy_signal"]].to_string())
+
+print("\n=== Around 2025-01-02 15:40-15:55 ===")
+sub = window[(window["timestamp"] >= "2025-01-02 15:40:00") & (window["timestamp"] <= "2025-01-02 15:55:00")]
+print(sub[["timestamp", "open", "high", "low", "close", "fast", "slow",
+          "rsi", "atr", "trend_up", "near_fast", "buy_cross", "buy_signal"]].to_string())