fix: chunk 配置通过 DocumentIngestor 传递，不再被硬编码覆盖

- DocumentIngestor 新增 chunk_config 参数，类型 ChunkConfig
- ingest_file 使用 self.chunk_config.max_size/overlap 代替硬编码 1000/100
- deps.py get_ingestor 传入 config.chunk 配置对象

Co-Authored-By: Claude <noreply@anthropic.com>

src/core/ingest.py

@@ -3,6 +3,7 @@ import logging
 import re
 from pathlib import Path
 
+from src.core.config import ChunkConfig
 from src.core.db import VectorDB
 from src.core.embedder import Embedder, batch_embed
 from src.core.splitters.markdown import MarkdownSplitter  # 兼容旧 import 路径
@@ -21,11 +22,13 @@ class DocumentIngestor:
         embedder: Embedder,
         collection_name: str,
         splitter: Splitter | None = None,
+        chunk_config: ChunkConfig | None = None,
     ):
         self.db = db
         self.embedder = embedder
         self.collection_name = collection_name
         self.splitter = splitter or MarkdownSplitter()
+        self.chunk_config = chunk_config or ChunkConfig()
 
     @property
     def collection(self):
@@ -43,7 +46,9 @@ class DocumentIngestor:
         file_name = f"{path_hash}_{path.name}"
 
         splitter = self.splitter or get_splitter(
-            file_path, max_size=1000, overlap=100
+            file_path,
+            max_size=self.chunk_config.max_size,
+            overlap=self.chunk_config.overlap,
         )
 
         # PDF/EPUB 二进制文件特殊处理：splitter 内部读取文件

src/server/deps.py

@@ -45,7 +45,10 @@ class AppState:
         name = collection or self.default_collection
         with self._cache_lock:
             if name not in self._ingestors:
-                self._ingestors[name] = DocumentIngestor(self.db, self.embedder, name)
+                self._ingestors[name] = DocumentIngestor(
+                    self.db, self.embedder, name,
+                    chunk_config=self.config.chunk,
+                )
             return self._ingestors[name]
 
     def list_collections_with_stats(self) -> list[dict]: