feat: enhance model training and deployment scripts with time-weighted sampling

- Updated `train_model.py` and `train_mlx_model.py` to include a time weight decay parameter for improved sample weighting during training.
- Modified dataset generation to incorporate sample weights based on time decay, enhancing model performance.
- Adjusted deployment scripts to support new backend options and improved error handling for model file transfers.
- Added new entries to the training log for better tracking of model performance metrics over time.
- Included ONNX model export functionality in the MLX filter for compatibility with Linux servers.

src/label_builder.py

@@ -9,21 +9,17 @@ def build_labels(
     stop_loss: float,
     side: str,
 ) -> Optional[int]:
-    """
-    진입 이후 미래 캔들을 순서대로 확인해 TP/SL 도달 여부를 판단한다.
-    LONG: high >= TP → 1, low <= SL → 0
-    SHORT: low <= TP → 1, high >= SL → 0
-    둘 다 미도달 → None (학습 데이터에서 제외)
-    """
     for high, low in zip(future_highs, future_lows):
         if side == "LONG":
-            if high >= take_profit:
-                return 1
+            # 보수적 접근: 손절(SL)을 먼저 체크
             if low <= stop_loss:
                 return 0
-        else:  # SHORT
-            if low <= take_profit:
+            if high >= take_profit:
                 return 1
+        else:  # SHORT
+            # 보수적 접근: 손절(SL)을 먼저 체크
             if high >= stop_loss:
                 return 0
+            if low <= take_profit:
+                return 1
     return None