Add GDCM dependency to Dockerfile

[danielsnider]

My DICOM images seem to be compressed and ask for GDCM. It would be great to add GDCM to the ocr Dockerfile :-).

This is what I see:

root@38a72d18ad9b:/code# ./entrypoint.sh --input /data
DEBUG Found 1 contender files in data
DEBUG Checking 1 dicom files for validation.
Found 1 valid dicom files
Traceback (most recent call last):
  File "/code/main.py", line 172, in <module>
    main()
  File "/code/main.py", line 108, in main
    verbose=args.verbose)
  File "/code/user/__init__.py", line 31, in __init__
    self.image = dicom._get_pixel_array()
  File "/opt/anaconda2/lib/python2.7/site-packages/pydicom-1.0.0a1-py2.7.egg/pydicom/dataset.py", line 1053, in _get_pixel_array
    self._pixel_array = self._pixel_data_numpy()
  File "/opt/anaconda2/lib/python2.7/site-packages/pydicom-1.0.0a1-py2.7.egg/pydicom/dataset.py", line 690, in _pixel_data_numpy
    raise NotImplementedError("Pixel Data is compressed in a "
NotImplementedError: Pixel Data is compressed in a format pydicom does not yet handle. Cannot return array. Pydicom might be able to convert the pixel data using GDCM if it is installed.

Sidenote: Commenting on line main.py#L157... if you catch all exceptions it would be great to at minimum print the exception instead of hiding it. Not seeing the error is the worst kind of error.