refactor(archive): move unused modules to archive/

Archive legacy framework and utility modules that are no longer
referenced by the active core (datasource/ and rotation/):

- framework/ -> archive/framework/
- framework_v2/ -> archive/framework_v2/
- strategies/ -> archive/strategies/
- config/ -> archive/config/
- visualization/ -> archive/visualization/
- scripts/ -> archive/scripts/
- tests/ -> archive/tests/
- run_rotation.py, run_us_rotation.py -> archive/single_files/
- compare_*.py, test_api_dates.py -> archive/single_files/

archive/framework_v2/shared/factors/__init__.py

@@ -0,0 +1,17 @@
+"""
+通用因子实现
+"""
+
+from framework_v2.shared.factors.momentum import MomentumFactor
+
+# TALibFactorBase 需要安装 talib，可选导入
+try:
+    from framework_v2.shared.factors.talib_base import TALibFactorBase
+    __all__ = [
+        'TALibFactorBase',
+        'MomentumFactor',
+    ]
+except ImportError:
+    __all__ = [
+        'MomentumFactor',
+    ]

archive/framework_v2/shared/factors/momentum.py

@@ -0,0 +1,104 @@
+"""
+动量因子（通用版本）
+
+使用加权线性回归：得分 = 年化收益率 × R²
+
+与现有 MomentumFactor 对比验证：
+- 输入相同 → 输出应该相同
+"""
+
+import pandas as pd
+import numpy as np
+import math
+from framework_v2.core import FactorBase
+
+
+class MomentumFactor(FactorBase):
+    """
+    动量因子
+    
+    计算加权线性回归动量得分：
+    得分 = 年化收益率 × R²
+    
+    参数：
+    - n_days: 动量窗口（默认25）
+    - weighted: 是否加权（默认True）
+    - crash_filter: 是否启用崩盘过滤（默认True）
+    """
+    
+    name = "momentum"
+    category = "momentum"
+    
+    def __init__(
+        self,
+        n_days: int = 25,
+        weighted: bool = True,
+        crash_filter: bool = True
+    ):
+        super().__init__(n_days=n_days, weighted=weighted, crash_filter=crash_filter)
+        self.n_days = n_days
+        self.weighted = weighted
+        self.crash_filter = crash_filter
+    
+    def compute(self, data: pd.DataFrame) -> pd.Series:
+        """计算动量因子值"""
+        if 'close' not in data.columns:
+            raise ValueError("data must contain 'close' column")
+        
+        prices = data['close']
+        
+        if self.weighted:
+            factor_values = prices.rolling(self.n_days).apply(
+                lambda x: self._weighted_momentum_score(x.values),
+                raw=False
+            )
+        else:
+            factor_values = prices.pct_change(self.n_days)
+        
+        if self.crash_filter:
+            factor_values = self._apply_crash_filter(prices, factor_values)
+        
+        return factor_values
+    
+    def _weighted_momentum_score(self, prices: np.ndarray) -> float:
+        """计算加权动量得分（完全复制现有逻辑）"""
+        if len(prices) < 5:
+            return 0.0
+        
+        # 价格下界 clip，防止 log(0) 或 log(负数)
+        prices = np.clip(prices, 0.01, None)
+        y = np.log(prices)
+        
+        # 异常值检测
+        if np.any(np.isnan(y)) or np.any(np.isinf(y)):
+            return 0.0
+        
+        x = np.arange(len(y))
+        weights = np.linspace(1, 2, len(y))
+        
+        slope, intercept = np.polyfit(x, y, 1, w=weights)
+        annualized_returns = math.exp(slope * 250) - 1
+        
+        y_pred = slope * x + intercept
+        ss_res = np.sum(weights * (y - y_pred) ** 2)
+        ss_tot = np.sum(weights * (y - np.average(y, weights=weights)) ** 2)
+        r2 = 1 - ss_res / ss_tot if ss_tot > 0 else 0
+        
+        return annualized_returns * r2
+    
+    def _apply_crash_filter(self, prices: pd.Series, factor_values: pd.Series) -> pd.Series:
+        """崩盘过滤：连续3天跌>5%清零（完全复制现有逻辑）"""
+        result = factor_values.copy()
+        
+        for i in range(3, len(prices)):
+            r1 = prices.iloc[i] / prices.iloc[i-1]
+            r2 = prices.iloc[i-1] / prices.iloc[i-2]
+            r3 = prices.iloc[i-2] / prices.iloc[i-3]
+            
+            con1 = min(r1, r2, r3) < 0.95
+            con2 = (r1 < 1) and (r2 < 1) and (r3 < 1) and (prices.iloc[i] / prices.iloc[i-3] < 0.95)
+            
+            if con1 or con2:
+                result.iloc[i] = 0.0
+        
+        return result

archive/framework_v2/shared/factors/talib_base.py

@@ -0,0 +1,55 @@
+"""
+ta-lib 因子基类（通用）
+
+所有 ta-lib 因子继承此类，只需指定函数和参数
+"""
+
+import talib
+import pandas as pd
+import numpy as np
+from framework_v2.core import FactorBase
+
+
+class TALibFactorBase(FactorBase):
+    """
+    ta-lib 因子基类
+    
+    子类只需实现：
+    - name: 因子名称
+    - _talib_func: 返回 ta-lib 函数
+    """
+    
+    category = "technical"
+    
+    def __init__(self, period: int = 14, **params):
+        """
+        初始化
+        
+        Args:
+            period: 周期参数
+            **params: 其他参数
+        """
+        super().__init__(period=period, **params)
+        self.period = period
+    
+    def compute(self, data: pd.DataFrame) -> pd.Series:
+        """
+        计算因子值
+        
+        Args:
+            data: OHLCV 数据
+        
+        Returns:
+            因子值序列
+        """
+        close = data['close'].values.astype(float)
+        
+        # 调用子类指定的 ta-lib 函数
+        result = self._talib_func(close, timeperiod=self.period)
+        
+        return pd.Series(result, index=data.index, name=self.name)
+    
+    @property
+    def _talib_func(self):
+        """子类必须实现，返回 ta-lib 函数"""
+        raise NotImplementedError("Subclasses must implement _talib_func")