feat(evaluators): send authenticated execution context to runners api

[namrataghadi-galileo]

Summary

• Add Galileo execution_context to the existing POST /api/v1/scorers/invoke request.
• Map the authenticated Agent Control context to Galileo fields: namespace_key → organization_id, caller_id → user_id, trusted extension project_id → project_id, and supported target ID → run_id.
• Emit optional context values as null; preserve legacy requests when no runtime extensions are present.

Scope

• User-facing/API changes: Galileo scorer invocation includes the additive execution_context object.
• Internal changes: Context translation stays inside the Galileo evaluator package. run_id is set only for log_stream and agent_stream targets.
• Out of scope: Generic Agent Control models/auth, scorer-grant flows, record serialization/factory, and record normalization.

Risk and Rollout

• Risk level: low
• Rollback plan: Revert the Galileo package changes; callers without runtime context continue using the legacy request shape.

Testing

• Added or updated automated tests
• Ran Galileo package tests: 172 passed
• Ran Galileo Ruff: all checks passed
• Ran make check (package Makefile targets could not resolve hatchling because the configured package index returned HTTP 401; equivalent package-local test and Ruff commands passed from the existing virtualenv)
• Manually verified behavior

Checklist

• Linked issue/spec (if applicable)
• Updated docs/examples for user-facing changes
• Included any required follow-up tasks

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!