Testing of tool call usage by LLM augmented with the MCP

[dennisfeng200]

Purpose

• adds integration tests that verify whether the intended tool functions are called when given sensible user messages.
• e.g. if user asks "what kinds of projects are on AI Foundry Labs" we test if the list-projects tool is called.

Does this introduce a breaking change?

[ ] Yes
[x] No

Pull Request Type

What kind of change does this Pull Request introduce?

[ ] Bugfix
[ ] Feature
[ ] Code style update (formatting, local variables)
[ ] Refactoring (no functional changes, no api changes)
[ ] Documentation content changes
[x] Other... Please describe: adds integration tests

How to Test

• Get the code

git clone [repo-address]
cd [repo-name]
git checkout [branch-name]
uv sync --prerelease allow --group test

• ensure you have the correct env variables. You need to fill out the following 3 env variables with an Azure OpenAI endpoint:

AZURE_OPENAI_ENDPOINT=<placeholder>
AZURE_OPENAI_API_VERSION="2025-03-01-preview"
AZURE_OPENAI_DEPLOYMENT="gpt-4o"

• Test the code

cd tests
uv run --prerelease allow --group test pytest --runintegration

What to Check

Verify that the following are valid

• all tests pass. Might take 5+ min because it's calling LLM API.

Other Information

[dem108]

I had shared a few checkpoints offline - please follow up to update this PR. Thanks! @dennisfeng200 @tendau