Add configurable preprocessing pipelines

[cmccomb]

Summary

• add a dedicated settings::preprocessing module with a composable pipeline/step DSL and expose the new builders on supervised settings
• refactor the preprocessor to execute arbitrarily long pipelines, including a stateful standardization step, and update the docs/examples to describe the workflow
• expand the regression/classification tests to cover the new APIs and ensure pipelines work end-to-end

Testing

• cargo fmt
• cargo test
• cargo clippy --all-targets -- -D warnings -D clippy::pedantic

---

Codex Task