Support unicode file names in data search

[paulmueller]

(similar to DC-analysis/dclab#50)

Searching for data sets in directories containing non-ASCII characters fails:

Exception in thread Thread-9:
Traceback (most recent call last):
  File "/usr/lib/python2.7/threading.py", line 801, in __bootstrap_inner
    self.run()
  File "/home/paul/repos/ShapeOut/shapeout/gui/gaugeframe.py", line 79, in run
    res=result.get(timeout=.1)
  File "/usr/lib/python2.7/multiprocessing/pool.py", line 572, in get
    raise self._value
UnicodeEncodeError: 'ascii' codec can't encode character u'\xe1' in position 41: ordinal not in range(128)