Support multiple tool call functions in remote vLLM inference provider

[terrytangyuan]

🚀 Describe the new functionality needed

Currently, the remote vLLM inference provider only supports a single tool call function. For example, if you use this example: https://github.com/meta-llama/llama-stack-apps/blob/main/examples/agents/e2e_loop_with_client_tools.py, only the first function passed to client_tools argument in AgentConfig will be used.

💡 Why is this needed? What if we don't build it?

Users won't be able to use multiple tool call functions with agent.

Other thoughts

No response

[Shreyanand]

@terrytangyuan I tried running https://github.com/meta-llama/llama-stack-apps/blob/main/examples/agents/e2e_loop_with_client_tools.py with Ollama backend and it works with multiple tools. I'm setting up a local vLLM server and connect llama-stack with it to test this with vLLM backend. I'll update the thread when I can reproduce the bug.

[github-actions]

This issue has been automatically marked as stale because it has not had activity within 60 days. It will be automatically closed if no further activity occurs within 30 days.