Toolchain Notes

This project favors a small, explicit toolchain so the performance work stays visible:

• cargo fmt for formatting.
• cargo clippy --all-targets --all-features for static checks.
• cargo test for library validation.
• cargo run --example ... for targeted architectural demos.
• GitHub Actions for Linux, macOS, and Windows checks, Docker image validation, and pinned Redis client-library suites.
• Linux nightly AddressSanitizer (asan job) and Miri (miri job). Persistence, TCP, Lua, and mimalloc stay out of Miri. The parser property test is skipped under Miri (covered by cargo fuzz); Miri still checks consume bounds on a few garbage frames. Reproduce with scripts/run_sanitizers.sh.

When you start benchmarking, capture the exact compiler, kernel, and CPU model so throughput numbers remain comparable.