`azd ai agent run` fails in LangGraph Toolbox scenario due to missing `AZURE_AI_MODEL_DEPLOYMENT_NAME`

[Menghua1]

Description:
When testing Scenario 3: LangGraph Toolbox (Responses), azd ai agent run fails. The log shows: ValueError: AZURE_AI_MODEL_DEPLOYMENT_NAME environment variable must be set.

Steps to Reproduce:

1. mkdir langgraph-toolbox and cd langgraph-toolbox
2. Run azd ai agent init --manifest "https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/langgraph-toolbox/agent.manifest.yaml" and select location northcentralus.
3. Run azd up.
4. Run export TOOLBOX_ENDPOINT=$(azd env get-value TOOLBOX_AGENT_TOOLS_MCP_ENDPOINT).
5. Run azd ai agent run.

Environment:

• Azd version: azd version 1.26.0-beta.1-daily.6335418 (commit 84dd634326889128cfc0b605e3956cf515be42a8) (daily).
• AI Agent Extension version: 0.1.32-preview
• Manifest URL: https://github.com/microsoft-foundry/foundry-samples/blob/main/samples/python/hosted-agents/bring-your-own/responses/langgraph-toolbox/agent.manifest.yaml
• OS: Windows
• Location: northcentralus.

Expected Behavior:
azd ai agent run can start successfully.

@JeffreyCA and @rajeshkamal5050 for notification.

[JeffreyCA]

@Menghua1 can you additionally run export AZURE_AI_MODEL_DEPLOYMENT_NAME=$(azd env get-value AZURE_AI_MODEL_DEPLOYMENT_NAME) before azd ai agent run?