[Bugfix]: Fix bug when using default embed in vector retrieval and url document.

[ChenJiahaoST]

📌 PR Description

• Fix bug of retrieval when using default embed (not dict format) and urldocument.
• Add set/get state for Retriever Module so that it can be set as ServerModule.

🔍Related Issue

• Fix [Bug]: Bug with UrlDocument #738

✅ 变更类型 / Type of Change

• 修复 Bug / Bug fix (non-breaking change that fixes an issue)
• 新功能 / New feature (non-breaking change that adds functionality)
• 重构 / Refactor (no functionality change, code structure optimized)
• 重大变更 / Breaking change (fix or feature that would cause existing functionality to change)
• 文档更新 / Documentation update (changes to docs only)

⚡ 更新后的用法示例 / Usage After Update

1. set document

docs = lazyllm.Document(dataset_path="", embed=embed_model, store_conf=store_conf,  server=True)
docs.create_node_group(name="line", display_name='句子切片', group_type=NodeGroupType.CHUNK, transform=lambda x: x.split("。"))
docs.activate_group("line")
docs.start()

2. set retriever

r = lazyllm.Retriever(docs, "block", 'cosine', topk=10)
r("query")
```
3. use server module
````python
from lazyllm import ServerModule
m = ServerModule(r)
m.start()
res = m("query")
```

## ⚠️ 注意事项 / Additional Notes
<!-- 是否有依赖更新、迁移步骤或其他注意点 / Mention dependencies, migration steps, or any other concerns -->
Note: Only when `server=True` (Document does have a url), retriever can be set as `ServerModule`