fix: Correct eval model_config and RBAC pass-through for CD pipeline

[ericchansen]

• Fix GroundednessEvaluator model_config validation failure by adding
  azure_deployment and credential params (was missing both)
• Extract base AI Services endpoint from project endpoint URL
  (evaluators need account-level endpoint, not project-scoped)
• Add configurable evaluation.model field to per-env agent configs
• Pass deployerPrincipalId to foundryAccount Bicep module (was missing,
  preventing RBAC assignment for pipeline identity)
• Update config SCHEMA.md with new evaluation.model field

Co-authored-by: Copilot 223556219+Copilot@users.noreply.github.com

[Copilot]

Pull request overview

This PR fixes evaluation configuration for real Foundry evaluator runs and passes the pipeline deployer identity into the Foundry account module for RBAC assignment.

Changes:

• Adds base AI Services endpoint extraction and configurable evaluator model wiring.
• Passes deployerPrincipalId into the Foundry account Bicep module.
• Documents and sets evaluation.model across environment configs.

Reviewed changes

Copilot reviewed 6 out of 6 changed files in this pull request and generated 1 comment.

Show a summary per file

File	Description
src/scripts/run_evaluation.py	Adds evaluator endpoint/model configuration for real evaluations.
infra/main.bicep	Passes deployer principal ID to the Foundry account module.
config/SCHEMA.md	Documents the new optional evaluation.model field.
config/agent-config.dev.json	Sets the evaluation scoring model for dev.
config/agent-config.test.json	Sets the evaluation scoring model for test.
config/agent-config.prod.json	Sets the evaluation scoring model for prod.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.