duplicate PixelData entry in report

[rkm]

I'm not quite sure how this occurred, but I've found an instance where running DicomFileRunner.ValidateDicomFile on a given file produces different results. Specifically, a duplicate FailurePart is returned for the PixelData. The only difference I can immediately spot is that the reports were produced from different machines however, both use the same software installation.

I've run out of time trying to reproduce this bug for the moment, so I'm opening this issue in case anyone else stumbles upon it.

If we can't fix it (e.g., due to a bug in the Tesseract lib we can't easily patch), a solution could be removing duplicate items before passing the report downstream.