feat: Add eval-harness for behavioral regression testing via Promptfoo and DeepEval

[vinodvx]

Description

Adds eval-harness for behavioral regression testing — mock LLM/tools, JSON run output, assertions on content, llm_usage, and telemetry.

• Go runner + shared run_agent.sh
• Promptfoo (eval-harness/promptfoo/)
• DeepEval (eval-harness/deepeval/)
• CI job on PRs
• README updates

Test plan

• [*] go run ./eval-harness/runner
• [*] cd eval-harness/promptfoo && npx promptfoo eval -c config.yaml
• [*] cd eval-harness/deepeval && pytest test_agent.py -v

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to change)
• Documentation update
• Refactor / chore

Related issues

Closes #28

Checklist

• I have run make check
• I have run task examples:all
• I have run make tidy if I added or removed dependencies
• Commit messages follow conventional commits (e.g. feat:, fix:, docs:)
• I have added/updated tests for my changes
• Documentation is updated if needed

[codecov]

Codecov Report

❌ Patch coverage is 50.82781% with 297 lines in your changes missing coverage. Please review.

Files with missing lines	Patch %	Lines
eval-harness/runner/setup/load.go	0.00%	61 Missing ⚠️
eval-harness/runner/setup/mock_llm.go	0.00%	60 Missing ⚠️
eval-harness/runner/setup/config.go	0.00%	35 Missing ⚠️
eval-harness/runner/setup/agent.go	0.00%	29 Missing ⚠️
eval-harness/runner/main.go	0.00%	25 Missing ⚠️
eval-harness/runner/setup/mock_tool.go	0.00%	24 Missing ⚠️
internal/runtime/local/agent_loop.go	84.12%	19 Missing and 1 partial ⚠️
internal/runtime/temporal/agent_workflow.go	82.22%	15 Missing and 1 partial ⚠️
internal/runtime/base/utils.go	0.00%	10 Missing ⚠️
eval-harness/runner/output.go	0.00%	7 Missing ⚠️
... and 3 more

📢 Thoughts on this report? Let us know!