MEAI evaluation updates - May

[gewarren]

Describe the issue or suggestion

• Update the list of packages and evaluators listed on this page.
• Update this tutorial to use RelevanceEvaluator + CoherenceEvaluator (or any other set that seems appropriate) in place of the now [Experimental] RelevanceTruthAndCompletenessEvaluator.
• Add new tutorial around how to use the evaluators included as part of the new MEAI.Evaluation.Safety package. These Safety evaluators require some additional configuration related to Azure AI Foundry. The eval samples in the dotnet/ai-samples repo have already been updated to include examples for this - See RunningSafetyEvaluators.cs and RunningQualityAndSafetyEvaluatorsTogether.cs. Can use these samples as a starting point for the new tutorial.
• All packages except MEAI.Evaluation.Safety will soon be marked GA. Update docs to reflect this. - Tracked by MEAI packages are stable - remove preview/prerelease verbiage #46233.

---

Associated WorkItem - 430203