修复模型名称: qwen-vl-plus

qwen-vl-latest 不是有效模型名，正确名称是:
- qwen-vl-plus (推荐)
- qwen-vl-max

参考阿里云官方示例代码

config.example.yaml

@@ -22,7 +22,7 @@ vlm:
   # 或直接在这里填写:
   # api_key: your-api-key-here
   api_key: ""
-  model_name: qwen-vl-latest  # 模型名称
+  model_name: qwen-vl-plus  # qwen-vl-plus / qwen-vl-max
   max_tokens: 2048
   temperature: 0.7
 

src/config/settings.py

@@ -16,7 +16,7 @@ class VLMSettings:
     model_type: str = "bailian"  # bailian / qwen-vl / gpt-4v
     api_base: str = "https://dashscope.aliyuncs.com/compatible-mode/v1"  # 阿里云百炼
     api_key: str = ""
-    model_name: str = "qwen-vl-latest"  # 推荐 qwen-vl-latest
+    model_name: str = "qwen-vl-plus"  # qwen-vl-plus / qwen-vl-max
     max_tokens: int = 2048
     temperature: float = 0.7
 

src/vlm/qwen_vl.py

@@ -94,11 +94,10 @@ class BailianVLMClient:
                     可从环境变量 ALIBABA_CLOUD_API_KEY 或 DASHSCOPE_API_KEY 获取
                     或从 https://bailian.console.aliyun.com/ 获取
             model_name: 模型名称
-                - qwen-vl-latest (推荐，VL 理解)
-                - qwen-vl2-7b
-                - qwen-vl2-72b
-                - qwen2-vl-72b-instruct
+                - qwen-vl-plus (推荐，VL 理解)
+                - qwen-vl-max
                 - qwen2.5-vl-72b-instruct
+                - qwen2.5-vl-32b-instruct
                 - qwen-omni-series (全模态)
             base_url: API 基础地址（OpenAI 兼容格式）
             max_tokens: 最大生成 token 数