strengthen Ruff linting and formatting

[laceyp99]

Summary

• upgrade the development toolchain to Ruff 0.16.2 and enable a broader curated set of correctness, modernization, testing, path, performance, and Ruff-native checks
• adopt Ruff's 88-character formatting baseline across evaluation orchestration, deterministic checks, analysis, and tests
• resolve the new lint findings with safer defaults, timezone-aware timestamps, explicit concurrency and iteration behavior, modern path handling, and focused test updates

Testing

• uv sync --locked --all-extras --python 3.12 - passed
• uv run --locked --all-extras --python 3.12 ruff format --check . - passed
• uv run --locked --all-extras --python 3.12 ruff check . - passed
• uv run --locked --all-extras --python 3.12 pytest -q - passed (130 tests)
• uv build - passed (source distribution and wheel)

[laceyp99]

Looks good to me! This is a cross project effort to make line length formatting uniform and have project relevant lint rules. This seems like a ton of LOC changes, but its mainly two simple config changes.