getFileIDs and getChecksums command line clients must scale to many files

The current implementation will only scale if all pillar have nearly the same sets of files in nearly the same sequence. This is because elements are purged from memory when the fileID value has been received from all pillars.