test(cli): add unit tests for dashboard runs CLI commands

[aaight]

Summary

• Adds tests/unit/cli/dashboard/runs/runs.test.ts with 34 unit tests covering all 9 runs CLI commands (list, show, logs, llm-calls, llm-call, debug, trigger, retry, cancel)
• Follows the DashboardCommand pattern from tests/unit/cli/dashboard/projects/integration-credentials.test.ts — mocks loadConfig, createDashboardClient, and chalk; uses a makeClient() factory
• Uses vi.useFakeTimers() for the debug --analyze --wait polling tests

Test coverage per command

• list — all filter flags (project, status, agent-type, limit, offset, sort, order), default values, pagination message, no-pagination case
• show — passes run ID to getById.query, JSON output
• logs — passes run ID, string/array output, "No logs found" message
• llm-calls — passes run ID, JSON output
• llm-call — passes run ID + call number, JSON output
• debug (show) — fetches analysis, handles null with helpful message, JSON output for both null and real analysis
• debug (trigger) — triggers analysis without waiting, JSON result flag
• debug (trigger+wait) — polls until completed, stops on idle, logs timeout after 5-min deadline with vi.useFakeTimers()
• trigger — passes project + agent-type, optional work-item-id + model, requires required flags, JSON output
• retry — passes run ID, optional model override, JSON output
• cancel — passes run ID, default/custom reason, JSON output

Test plan

• npx vitest run --project unit-core tests/unit/cli/dashboard/runs/ — 34 tests pass
• npx biome check tests/unit/cli/dashboard/runs/runs.test.ts — no lint errors

Trello card: https://trello.com/c/7qP5NXnK/458-as-a-developer-i-want-unit-tests-for-dashboard-runs-cli-commands-so-that-run-management-flows-are-covered

🤖 Generated with Claude Code

🕵️ claude-code · claude-sonnet-4-6 · run details

[nhopeatall]

LGTM — Well-structured unit tests that follow the established DashboardCommand test pattern from integration-credentials.test.ts. All 9 runs CLI commands are covered with correct tRPC client mocking, flag/argument mapping verified against the source implementations, and the vi.useFakeTimers() approach for the debug polling tests is sound. CI is green.

🕵️ claude-code · claude-opus-4-6 · run details

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!