fix: Fix converters/loaders, bump to v0.2.2, migrate to factslab org

[aaronstevenwhite]

Description

Rewrites and fixes converters/loaders for all four resources, bumps the version to 0.2.2, migrates the repository from aaronstevenwhite/glazing to factslab/glazing, and adds CI/CD and GitHub templates.

Type of Change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation update

Key Changes

• WordNet: Rewrote converter and loader to use enriched single-file JSONL with supplementary sense and exception files
• FrameNet: Added frame relation, LU enrichment, semantic type, and fulltext parsing to converter and loader
• VerbNet: Fixed converter to populate framenet_mappings and propbank_mappings from member attributes
• PropBank: Added AMR-UMR-91 roleset conversion and XML fixes
• Validation: Relaxed lemma validation to allow uppercase, leading digits, and dots
• Infrastructure: Added PyPI publish workflow (trusted publishers/OIDC), PR template, and issue templates
• Repo migration: Updated all URLs and references from aaronstevenwhite to factslab
• Version bump: 0.2.1 → 0.2.2
• Tests: Added converter-to-loader round-trip integration tests for all four resources

Impact

Users must reconvert datasets after upgrading (glazing init --force) to pick up the enriched JSONL output.

Testing

• All tests pass (pytest)
• New tests added (if applicable)
• Type checking passes (mypy --strict src/)
• Linting passes (ruff check)
• Formatting passes (ruff format)