重构项目结构并更新README.md

1. 重构目录结构：
   - 创建src/visualization模块用于存放可视化相关功能
   - 移动script/visualize_csv_comparison.py到src/visualization/comparison.py
   - 创建src/visualization/__init__.py导出主要函数
   - 整理script目录，按功能分类存放脚本文件

2. 更新README.md：
   - 添加CSV文件比较可视化部分
   - 提供Python API和命令行使用方法说明
   - 描述功能特点和使用示例

3. 更新模块引用：
   - 修正comparison.py中的模块引用路径
   - 更新命令行帮助信息中的使用示例

script/data_processing/merge_splits.py

@@ -0,0 +1,70 @@
+#!/usr/bin/env python3
+"""Merge split CSVs into a single file with split source labels."""
+
+import argparse
+from pathlib import Path
+
+import pandas as pd
+
+# Mapping of split CSV filenames to numeric labels for the source column
+SPLIT_LABELS = {
+    "split_test.csv": 1,
+    "split_train.csv": 2,
+    "split_val.csv": 3,
+}
+
+DEFAULT_COLUMN_NAME = "split_source"
+
+
+def parse_args() -> argparse.Namespace:
+    repo_root = Path(__file__).resolve().parent.parent
+    parser = argparse.ArgumentParser(
+        description=
+        "Combine split_*.csv files from splits_v2 and label their origin with integers."
+    )
+    parser.add_argument(
+        "--input-dir",
+        type=Path,
+        default=repo_root / "splits_v2",
+        help="Directory containing split_*.csv files (default: %(default)s)",
+    )
+    parser.add_argument(
+        "--output",
+        type=Path,
+        default=repo_root / "data" / "merged_splits.csv",
+        help="Destination CSV path (default: %(default)s)",
+    )
+    parser.add_argument(
+        "--column-name",
+        default=DEFAULT_COLUMN_NAME,
+        help="Name for the source column (default: %(default)s)",
+    )
+    return parser.parse_args()
+
+
+def main() -> None:
+    args = parse_args()
+
+    if not args.input_dir.is_dir():
+        raise SystemExit(f"Input directory not found: {args.input_dir}")
+
+    frames = []
+    for filename, label in SPLIT_LABELS.items():
+        csv_path = args.input_dir / filename
+        if not csv_path.is_file():
+            raise SystemExit(f"Missing expected split file: {csv_path}")
+        df = pd.read_csv(csv_path)
+        df[args.column_name] = label
+        frames.append(df)
+
+    if not frames:
+        raise SystemExit("No split CSV files were loaded.")
+
+    merged = pd.concat(frames, ignore_index=True)
+    args.output.parent.mkdir(parents=True, exist_ok=True)
+    merged.to_csv(args.output, index=False)
+    print(f"Merged {len(frames)} files with {len(merged)} rows into {args.output}")
+
+
+if __name__ == "__main__":
+    main()