sort out duplicate files

we have some filename case duplicates at least - eg gS20150111S0234_bias.fits.bz2 and uppercase filename.
Add a duplicate column to tapefile (default to false) and then mark the upper case ones as duplicates. Update request_from_tape to ignore duplicates.

Write a script to delete the duplicates from the archive - set diskfile.present to False and actually delete the dupe from S3