Updated code linting reports

[robgruen]

This pull request introduces several improvements and cleanups to the code quality tooling and tests, focusing on making code debt gating more robust and maintainable. It adds support for baseline exceptions in code quality gates, improves skipped/focused test detection, re-enables a previously skipped grammar test, and replaces a dead test with a more meaningful one. There are also small updates to documentation and npm scripts to reflect these changes.

Code quality tooling enhancements:

• Added support for baseline exception files (--exceptions-file) to the code-lint, code-complexity, code-debt, and code-circular tools, allowing known issues to be grandfathered and not block the ratchet/gate. Exception file format and usage are documented in README.md. [1] [2] [3] [4] [5] [6] [7]
• Improved detection of skipped and focused tests in code-debt tooling: now recognizes .skip.each and .only.each, and ignores placeholder stubs (empty () => {} bodies). [1] [2] [3]

Test and CI improvements:

• Re-enabled the CalendarDate grammar loader test in dynamicGrammarLoader.spec.ts now that the required converter exists, and updated its expected output to match the new slot value format. [1] [2]
• Removed the dead, no-op api/test/api.spec.ts and replaced it with a real web server smoke test in api.test.ts that exercises static file serving and 404 handling. [1] [2]
• Updated the code-lint and lint:fix npm scripts to use the new ratchet/gate mechanisms and only fix changed files. [1] [2]

Documentation updates:

• Updated README.md to reflect new test debt numbers (skipped tests down to 2, both intentional), improved explanations of the ratchet/gate process, and documented the new baseline exception mechanism. [1] [2] [3] [4]

Other minor improvements:

• Adjusted reporting in circularDepsReport.ts and complexityReport.ts to show all relevant results instead of slicing to a top-N, and improved output for baseline exceptions. [1] [2]

These changes collectively make code quality enforcement more flexible and accurate, while reducing false positives and technical debt noise.