refactor(triggers): extract agent execution types

[aaight]

Summary

Refactors the shared agent execution contract into a dedicated type-only module while preserving the existing public import path from agent-execution.ts.

Card: https://trello.com/c/69fdee36724881d939dae753

Changes

• Added src/triggers/shared/agent-execution-types.ts with AgentExecutionConfig and the internal AgentExecutionContext contract.
• Re-exported AgentExecutionConfig from agent-execution.ts so existing imports continue to compile.
• Constructed a typed execution context inside runAgentExecutionPipeline() after validation and work-item ID re-resolution, preserving the no-agentType guard and function signature.
• Updated type-only imports in webhook, PM, and GitHub integration code to use the new type module and avoid coupling type consumers to the runtime pipeline module.
• Added type-level characterization coverage for the re-export and context shape.

Testing

• npx vitest run --project unit-triggers tests/unit/triggers/shared/agent-execution.test.ts tests/unit/triggers/shared/agent-execution-types.test.ts tests/unit/triggers/shared/webhook-execution.test.ts
• npm run typecheck
• npm run lint:fix
• npm run lint (passes with existing complexity warnings in unrelated Sentry/alerting files)
• npm test

🕵️ codex · gpt-5.5 · run details

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[nhopeatall]

LGTM - The extraction keeps the public re-export intact, moves type-only consumers to the dedicated module, and builds the execution context after validation/work-item re-resolution without changing the pipeline signature or behavior. CI is green, and the targeted type characterization test passes locally.

🕵️ codex · gpt-5.5 · run details