Potential python 3.x issue in calling ambertools

[jchodera]

We may have a Python 3 / unicode issue:

cc: https://github.com/openeye-private/openmm_orion/issues/1#issuecomment-267092169

 File "/Users/jlafon/sources/openmm_orion/OpenMMCubes/cubes.py", line 96, in process
    modeller.addHydrogens(forcefield, self.args.pH)
  File "/Users/jlafon/anaconda/envs/openmm/lib/python3.5/site-packages/simtk/openmm/app/modeller.py", line 863, in addHydrogens
    system = forcefield.createSystem(newTopology, rigidWater=False, nonbondedMethod=CutoffNonPeriodic)
  File "/Users/jlafon/anaconda/envs/openmm/lib/python3.5/site-packages/simtk/openmm/app/forcefield.py", line 1067, in createSystem
    if generator(self, res):
  File "/Users/jlafon/anaconda/envs/openmm/lib/python3.5/site-packages/openmoltools/forcefield_generators.py", line 495, in gaffTemplateGenerator
    molecule = generateOEMolFromTopologyResidue(residue)
  File "/Users/jlafon/anaconda/envs/openmm/lib/python3.5/site-packages/openmoltools/forcefield_generators.py", line 139, in generateOEMolFromTopologyResidue
    [status, output] = getstatusoutput(command)
  File "/Users/jlafon/anaconda/envs/openmm/lib/python3.5/subprocess.py", line 805, in getstatusoutput
    data = check_output(cmd, shell=True, universal_newlines=True, stderr=STDOUT)
  File "/Users/jlafon/anaconda/envs/openmm/lib/python3.5/subprocess.py", line 626, in check_output
    **kwargs).stdout
  File "/Users/jlafon/anaconda/envs/openmm/lib/python3.5/subprocess.py", line 695, in run
    stdout, stderr = process.communicate(input, timeout=timeout)
  File "/Users/jlafon/anaconda/envs/openmm/lib/python3.5/subprocess.py", line 1059, in communicate
    stdout = self.stdout.read()
  File "/Users/jlafon/anaconda/envs/openmm/lib/python3.5/codecs.py", line 321, in decode
    (result, consumed) = self._buffer_decode(data, self.errors, final)
UnicodeDecodeError: 'utf-8' codec can't decode byte 0xfe in position 82064: invalid start byte