Response API: Support non vector store assistants

[AkhileshNegi]

Summary

Target issue is #241

Checklist

Before submitting a pull request, please ensure that you mark these task.

• Ran fastapi run --reload app/main.py or docker compose up in the repository root and test.
• If you've fixed a bug or added code that is tested and has test cases.

Notes

• Updated code
• Added testcases

[codecov]

Codecov Report

Attention: Patch coverage is 97.43590% with 1 line in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
backend/app/tests/api/routes/test_responses.py	97.05%	1 Missing ⚠️

📢 Thoughts on this report? Let us know!

[avirajsingh7]

@AkhileshNegi We can refactor this

params = {
    "model": assistant.model,
    "previous_response_id": request.response_id,
    "instructions": assistant.instructions,
    "temperature": assistant.temperature,
    "input": [{"role": "user", "content": request.question}],
}

if assistant.vector_store_id:
    params["tools"] = [
        {
            "type": "file_search",
            "vector_store_ids": [assistant.vector_store_id],
            "max_num_results": assistant.max_num_results,
        }
    ]
    params["include"] = ["file_search_call.results"]

response = client.responses.create(**params)

[AkhileshNegi]

done, thanks for sharing