fix(cli): improve quickstart e2e failure handling

[BloggerBust]

Description

Improves the quickstart e2e test so trigger and runtime failures fail CI instead of being logged as warnings. Previously, a generated project could fail to run end-to-end while the test still passed.

This change fails the test when the workflow trigger command fails, finds worker runner errors before triggering, and reports worker process errors with clearer context after cancellation. The quickstart trigger commands use synchronous SDK calls that wait for workflow completion. A successful trigger verifies that the generated worker processed the workflow rather than only enqueueing it.

Fixes # None

Type of change

• Bug fix (non-breaking change which fixes an issue)
• CI (any automation pipeline changes)
• Test changes (add, refactor, improve or change a test)

What's Changed

• Fail quickstart e2e tests when the workflow trigger command fails.
• Check for worker runner errors before waiting on the workflow result.
• Improve worker process error reporting after cancellation.

Checklist

Changes have been:

• Tested (unit, integration, or manually with steps specified)
• Linted and formatted
• Documented (where applicable)
• Added to CHANGELOG (where applicable) -- see [Keep a Changelog](https://keepachangelog.com/en/1.0.0/)

Testing

Validated locally with:

go vet ./cmd/hatchet-cli/cli/...
go build ./cmd/hatchet-cli/cli/...
go build -tags e2e_cli ./cmd/hatchet-cli/cli/...

Full quickstart e2e execution will be further tested by the CI for this PR.

---

🤖 AI Disclosure

• I acknowledge that an LLM was used in the creation of this Pull Request, in accordance with Hatchet's AI_POLICY.md.

• Details: Used Claude Code to explain failure modes and verify behavior.

[vercel]

@BloggerBust is attempting to deploy a commit to the Hatchet Team on Vercel.

A member of the Team first needs to authorize it.