Build a sorted subject index that can be reused across repeated overlap queries.
Details
The index stores a sorted representation of subject that is optimized for
repeated overlap queries.
Build the index once, then pass it as subject to
fast_find_overlaps(), fast_count_overlaps(), or other overlap-summary
functions.
Use the raw subject object, not the index, when you need subject metadata
in the output table, for example with overlap joins.