Takes a sequence of Path objects (e.
Takes a sequence of Path objects (e.g. the return value of findFiles). Treats each path as corresponding to a Read set -- loads each Read set, converts each set to use the same SequenceDictionary, and returns the union of the RDDs.
(GenomeBridge is using this to load BAMs that have been split into multiple files per sample, for example, one-BAM-per-chromosome.)
The locations of the parquet files to load
a single RDD[Read] that contains the union of the ADAMRecords in the argument paths.