feat(training): TabM模型量化交易优化

- 新增 CrossSectionSampler 支持截面数据采样（按交易日批处理）
- 新增 EnsembleQuantLoss (Huber + IC) 替代 MSE 作为损失函数
- 重构 TabMModel 支持量化场景：Rank IC 作为验证指标、CosineAnnealingLR学习率调度、梯度裁剪
- 支持 date_col 参数和特征对齐
- 更新实验配置 batch_size 2048 和 weight_decay 等超参数

src/training/components/models/__init__.py

@@ -7,5 +7,14 @@ from src.training.components.models.lightgbm import LightGBMModel
 from src.training.components.models.lightgbm_lambdarank import LightGBMLambdaRankModel
 from src.training.components.models.tabpfn_model import TabPFNModel
 from src.training.components.models.tabm_model import TabMModel
+from src.training.components.models.cross_section_sampler import CrossSectionSampler
+from src.training.components.models.ensemble_quant_loss import EnsembleQuantLoss
 
-__all__ = ["LightGBMModel", "LightGBMLambdaRankModel", "TabPFNModel", "TabMModel"]
+__all__ = [
+    "LightGBMModel",
+    "LightGBMLambdaRankModel",
+    "TabPFNModel",
+    "TabMModel",
+    "CrossSectionSampler",
+    "EnsembleQuantLoss",
+]