update: use nightly sglang

docs/sglang_deploy_guide.md

@@ -40,7 +40,9 @@ We recommend installing SGLang in a fresh Python environment:
 ```bash
 uv venv
 source .venv/bin/activate
-uv pip install "sglang" --prerelease=allow
+git clone https://github.com/sgl-project/sglang
+cd sglang
+uv pip install -e "python" --prerelease=allow
 ```
 
 Run the following command to start the SGLang server. SGLang will automatically download and cache the MiniMax-M2.1 model from Hugging Face.

docs/sglang_deploy_guide_cn.md

@@ -40,7 +40,9 @@
 ```bash
 uv venv
 source .venv/bin/activate
-uv pip install "sglang" --prerelease=allow
+git clone https://github.com/sgl-project/sglang
+cd sglang
+uv pip install -e "python" --prerelease=allow
 ```
 
 运行如下命令启动 SGLang 服务器，SGLang 会自动从 Huggingface 下载并缓存 MiniMax-M2.1 模型。