Python: Replace Eval SDK with AI Projects SDK in evaluation sample

[YusakuNo1]

Motivation and Context

Description

Contribution Checklist

• The code builds clean without any errors or warnings
• The PR follows the Contribution Guidelines
• All unit tests pass, and I have added new tests where possible
• Is this a breaking change? If yes, add "[BREAKING]" prefix to the title of the PR.

[Copilot]

Pull request overview

This PR replaces the Azure AI Evaluation SDK with the Azure AI Projects SDK in the self-reflection evaluation sample, transitioning from the GroundednessEvaluator to the OpenAI evals API for groundedness evaluation.

Key Changes:

• Migrated from azure-ai-evaluation SDK to azure-ai-projects SDK with OpenAI client integration
• Replaced GroundednessEvaluator with direct OpenAI evals API calls (client.evals.create() and client.evals.runs.create())
• Updated environment variables from AZURE_OPENAI_ENDPOINT/AZURE_OPENAI_API_KEY to AZURE_AI_PROJECT_ENDPOINT

Reviewed changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 9 comments.

File	Description
python/samples/getting_started/evaluation/self_reflection/self_reflection.py	Replaced evaluation SDK with AI Projects SDK; refactored evaluation logic to use OpenAI evals API with polling mechanism
python/samples/getting_started/evaluation/self_reflection/README.md	Updated documentation to reflect JSONL format, new SDK dependency, and new environment variable requirements
python/samples/getting_started/evaluation/self_reflection/.env.example	Updated environment variable template from OpenAI endpoint/key to AI Project endpoint

Comments suppressed due to low confidence (1)

python/samples/getting_started/evaluation/self_reflection/self_reflection.py:189

• Testing for None should use the 'is' operator.

        if eval_run_output_items == None:

---

You can also share your feedback on Copilot code review for a chance to win a $100 gift card. Take the survey.