BUG: Fixed randomizing order of dropped items in subject hierarchy

If multiple items were selected, and drag&dropped under a new parent, the order of the items under the new parent was randomized (to reproduce it was enough to load several MRHeads, drop them under a study in one move, then invoking DICOM export on the study - the items were in random order). Now it is fixed, and the order is preserved. Code complexity also decreased slightly.

Also style fix in terminology logic, and minor fix in DICOM export dialog UI

git-svn-id: http://svn.slicer.org/Slicer4/trunk@25895 3bd1e089-480b-0410-8dfb-8563597acbee