[pygwalker] Larger dataset supports based on duckDB

New feature: pygwalker supports larger datasets by running all OLAP computations on DuckDB locally.

In the past, pygwalker ran all computations in Webworker with javascript, meaning there is a data limit to be handled (v8 memory limit). This update upscales the limit to your local devices.

PyGWalker + DuckDB is all you need for large data exploration