Improve doctor.mjs test suite based on review feedback

[Copilot]

The original doctor.test.mjs had structural issues: repeated await import() calls per-test, incorrect test naming, informal comments, missing coverage for several code paths, and no tests for the verbose flag.

Changes

Mock setup refactored

• validateSpec imported at top level; default mock set in beforeEach instead of per-test dynamic imports
• vi.mock('fs') retained — required in ESM to make named exports configurable for vi.spyOn

Test quality fixes

• Renamed test: "should report error if overlay settings file is missing" → "should report warning ..." (matches actual behavior)
• Removed informal/uncertain inline comment

New coverage

• Unknown render targets are silently ignored (no spurious findings)
• Success path when all configured template roots exist
• project.yaml invalid YAML triggers error finding (previously untested catch block)
• project.yaml completeness test now asserts both the info percentage finding and the Top missing high-impact fields warning

Verbose flag tests (3 new)

• Asserts flags.verbose prints suggested next actions for each status (FAIL, WARN, PASS)

Refactor

• Extracted getConsoleOutput() helper to eliminate duplication across verbose tests
• Renamed findingWithUnknownTarget → unknownTargetFinding

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.