We should simplify the process to use KDB-Tree partitioner to generate geoparquet files

[jiayuasu]

Right now, it requires multiple steps to get it, including StructuredAdapter

[jiayuasu]

After giving this some thoughts, I don't think this is worth the effort. Existing approach such as sorting by geohash is already good enough. KDB-Tree partition also does not guarantee zero overlaps among the spatial extents of each file.