[Bug]: glm-4.7作为planning模型会请求报错400

### Version ```sh system:macos version: latest ``` ### Details 这个是我的配置.env ``` MIDSCENE_MODEL_BASE_URL="https://open.bigmodel.cn/api/paas/v4" MIDSCENE_MODEL_API_KEY="...." MIDSCENE_MODEL_NAME="glm-4.6v" MIDSCENE_MODEL_FAMILY="glm-v" MIDSCENE_PLANNING_MODEL_API_KEY="和上面一样" MIDSCENE_PLANNING_MODEL_BASE_URL="https://open.bigmodel.cn/api/paas/v4" MIDSCENE_PLANNING_MODEL_NAME="glm-4.7" ``` 这是报错 ``` [2026-03-20T15:23:29.448+08:00] AI call failed (attempt 1/2), retrying in 2000ms... Error: 400 API 调用参数有误，请检查文档。 [2026-03-20T15:23:32.613+08:00] call AI error BadRequestError: 400 API 调用参数有误，请检查文档。 at APIError.generate (node_modules/openai/core/error.mjs:41:20) at OpenAI.makeStatusError (node_modules/openai/client.mjs:162:32) at OpenAI.makeRequest (node_modules/openai/client.mjs:330:30) at process.processTicksAndRejections (node:internal/process/task_queues:105:5) at async callAI (node_modules/@midscene/core/dist/es/ai-model/service-caller/index.mjs:276:32) at async plan (node_modules/@midscene/core/dist/es/ai-model/llm-planning.mjs:157:62) at async Object.executor (node_modules/@midscene/core/dist/es/agent/tasks.mjs:140:38) at async TaskRunner.flush (node_modules/@midscene/core/dist/es/task-runner.mjs:141:35) at async TaskExecutor.runAction (node_modules/@midscene/core/dist/es/agent/tasks.mjs:112:28) at async runAiAct (node_modules/@midscene/core/dist/es/agent/agent.mjs:318:46) { status: 400 ``` ### Reproduce link no link ### Reproduce Steps node main.js