OK, pretty sure I know what the problem is. It doesn't have anything to do with the files being sparse per se. The problem is that when you converted the files to sparse, it changed the "last modified time" for each file, but didn't change them in any other way (i.e. the contents of the files are still the same.) This exposed a bug in disParity.
The bug is that when the "last modified time" of a file changes, disParity notices that fact during a scan and re-reads it to compute the checksum to see if the contents actually changed. If the contents changed, it removes it and re-adds it during the Update and everything is fine. The bug is that if the file hasn't actually changed, it doesn't update its saved value for the last modified time. So on each subsequent scan, it keeps thinking the last modified time has changed, so it keeps re-checking the checksum on all those files every time, which is why the scans are now taking so long.
I'm hesitant to put out a new version to everyone right now to fix this tiny bug, but it will certainly be fixed in the next release. In the meantime, you could fix it by removing all the affected files off of the drive, doing an update, then adding them back and doing another update. Or you could remove/re-add the entire drive if that would be easier.