Bundle Ruff for Python codegen formatting

[hardbyte]

Summary

• Format generated Python with a bundled dprint Ruff wasm artifact instead of shelling out to a local ruff binary
• Add a small wasmi-based host for the dprint v3 formatter ABI with checksum validation
• Update CLI tests/docs/CI now that Python codegen no longer depends on ruff being on PATH

Validation

• cargo fmt --all
• SQLX_OFFLINE=true cargo clippy --all-targets --all-features -- -D warnings
• SQLX_OFFLINE=true cargo build --workspace
• cargo test -p reflectapi python_formatter --features codegen
• cargo test -p reflectapi-cli python_format
• cargo run -p reflectapi-cli --quiet -- codegen --language python --schema reflectapi-demo/reflectapi.json --output /tmp/opencode/reflectapi-wasm-python-final --python-package-name api_client --python-sync --format=true && ruff format --check /tmp/opencode/reflectapi-wasm-python-final

Notes

• cargo test was also run; unrelated demo run snapshots failed due JSON field ordering drift in run__json/run__msgpack, so no snapshot updates were included.

[claude]

Claude Code Review

This repository is configured for manual code reviews. Comment @claude review to trigger a review and subscribe this PR to future pushes, or @claude review once for a one-time review.

_{Tip: disable this comment in your organization's Code Review settings.}

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: 3338920c38

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Reuse the bundled formatter across generated files

For package/directory generation, render_python_file is called once per module plus support files, but each call reaches this function and recompiles/instantiates the 4 MB Ruff Wasm module (and re-hashes it) from scratch. Schemas with many namespaces will therefore pay the full Wasm startup cost hundreds of times during one codegen -l python -f run; keep one formatter/module for the generation pass or cache the compiled module so formatting scales with file count rather than repeated Wasm compilation.

Useful? React with 👍 / 👎.

[github-actions]

📖 Documentation Preview: https://reflectapi-docs-preview-pr-166.partly.workers.dev

Updated automatically from commit aa1dee1