feat: implement issue #1973 — [Engine config · Phase 1] Versioned config/ai-engines.json as the single source of truth for providers, model catalog and per-task chains
Problem
[Engine config · Phase 1] Versioned config/ai-engines.json as the single source of truth for providers, model catalog and per-task chains
From the issue: As an operator of the review and dev-lead engines, I want one versioned, schema-validated file that says which providers are enabled, which models exist, and what each task's default and fallback models are, so that model changes are reviewed, canaried with the release channels, and caught at PR time, instead of spread over about 17 Actions variables and hard-coded strings.
Risk
Medium — changes GitHub Actions workflow behavior, which is exercised only post-merge; verify via the affected workflow runs.
Test plan
Tests added/updated: tests/dev-lead/unit/test_ai_engines_config.bats,tests/test_batch_fallback.bats tests/test_consumer_manifest.bats,tests/test_validate_ai_engines.bats. Verification: bash scripts/dev-lead-lint.sh (shellcheck --severity=warning) ran pre-commit; the bats suite runs in CI.
Rollback
Revert this PR. No non-revertible side effects (no tags, migrations, or external state).
Monitoring
Watch the affected workflow run(s) in the Actions tab and this PR's Lint check for regressions.
Closes #1973