Consider removing ruff from linting pipeline

[dandye]

Background

Currently we use both ruff (for linting) and pyink (for formatting) in our pre-commit pipeline. Since we switched to pyink for formatting to support 2-space indentation, we should evaluate whether ruff is still providing enough value to justify the complexity.

Current Setup

• ruff: Python linting (pycodestyle, pyflakes, etc.)
• pyink: Python formatting with 2-space indentation
• mypy: Type checking
• bandit: Security scanning

Considerations

Pros of keeping ruff:

• Very fast (Rust-based)
• Comprehensive linting rules
• Active development
• Industry standard

Cons of keeping ruff:

• Additional tool to configure and maintain
• Some overlap with pyink's style checking
• Added complexity for developers

Options to Evaluate

1. Keep current setup - ruff + pyink + mypy
2. Remove ruff entirely - rely on pyink + mypy + bandit
3. Minimal ruff config - keep only basic error checking (F, E, W codes)
4. Replace with flake8 - more traditional, potentially simpler

Success Criteria

• Maintain code quality standards
• Minimize developer friction
• Keep CI/CD pipeline efficient
• Ensure 2-space indentation is enforced

Next Steps

• Benchmark current linting performance
• Test pyink-only workflow on a branch
• Gather developer feedback
• Make decision and implement