fix(config): allow Milvus vector store type in TOML configuration (#3041)

When users set `type = "milvus"` in their TOML vector storage config,
the server fails to start with:
  ValueError: Unknown type value: milvus, known types: [...]

Root causes:
1. The `StorageConfig.vector` field was annotated as `ChromaVectorConfig`
   instead of the base `VectorStoreConfig`, preventing polymorphic type
   dispatch from resolving any non-chroma vector store type correctly.
2. `milvus_store.py` imported `pymilvus` at module level. When pymilvus
   is not installed, the entire module fails to load and
   `MilvusVectorConfig` is never registered in the type registry.

Fix:
- Change the `vector` field type annotation to `VectorStoreConfig`
  (keeping `ChromaVectorConfig()` as default for backward compatibility).
- Move the top-level `pymilvus` import into the methods that actually
  use it (lazy import), matching the pattern used by all other vector
  store implementations (chroma, elasticsearch, pgvector, etc.).

Co-authored-by: Cursor <cursoragent@cursor.com>

packages/dbgpt-app/src/dbgpt_app/config.py

@@ -8,6 +8,7 @@ from dbgpt.model.parameter import (
     ModelServiceConfig,
 )
 from dbgpt.storage.cache.manager import ModelCacheParameters
+from dbgpt.storage.vector_store.base import VectorStoreConfig
 from dbgpt.util.configure import HookConfig
 from dbgpt.util.i18n_utils import _
 from dbgpt.util.parameter_utils import BaseParameters
@@ -55,7 +56,7 @@ class SystemParameters:
 class StorageConfig(BaseParameters):
     __cfg_type__ = "app"
 
-    vector: Optional[ChromaVectorConfig] = field(
+    vector: Optional[VectorStoreConfig] = field(
         default_factory=lambda: ChromaVectorConfig(),
         metadata={
             "help": _("default vector type"),

packages/dbgpt-ext/src/dbgpt_ext/storage/vector_store/milvus_store.py

@@ -9,8 +9,6 @@ import re
 from dataclasses import dataclass, field
 from typing import Any, Iterable, List, Optional
 
-from pymilvus.milvus_client import IndexParams, MilvusClient
-
 from dbgpt.core import Chunk, Embeddings
 from dbgpt.core.awel.flow import Parameter, ResourceCategory, register_resource
 from dbgpt.storage.vector_store.base import (
@@ -289,6 +287,14 @@ class MilvusStore(VectorStoreBase):
             connect_kwargs["user"] = self.username
             connect_kwargs["password"] = self.password
 
+        try:
+            from pymilvus.milvus_client import MilvusClient
+        except ImportError:
+            raise ValueError(
+                "Could not import pymilvus python package. "
+                "Please install it with `pip install pymilvus`."
+            )
+
         url = f"http://{self.uri}:{self.port}"
         self._milvus_client = MilvusClient(
             uri=url, user=self.username, db_name="default"
@@ -374,6 +380,8 @@ class MilvusStore(VectorStoreBase):
             )
             schema.add_function(bm25_fn)
         # Create the collection
+        from pymilvus.milvus_client import IndexParams
+
         collection = Collection(collection_name, schema)
         self.col = collection
         index_params = IndexParams()