Microoptimization for singlebyte fonts (#84)

Instead of list comprehension which will call a function to get the integer value of the bytes directly convert it to bytearray which is more optimal structure for storing list of bytes.

pdfminer/pdffont.py

@@ -494,7 +494,7 @@ class PDFFont(object):
         return False
 
     def decode(self, bytes):
-        return [six.indexbytes(bytes, i) for i,_ in enumerate(bytes)] # map(ord, bytes)
+        return bytearray(bytes)  # map(ord, bytes)
 
     def get_ascent(self):
         return self.ascent * self.vscale