improve handling of unicode and str in python2 code when using futurize

[jogo]

If python2 code uses unicode and str (to represent bytes).

When using futurize we add from builtins import str which breaks the behavior of the old str usage.

It would be nice to make it harder to mix up the two types when using futurize.

[jogo]

Before

print(str('a')==bytes('a'))
unicode('b')

This prints True

futurize -0 output

@@ -1,2 +1,5 @@
-print str('a')==bytes('a')
-unicode('b')
+from __future__ import print_function
+from builtins import bytes
+from builtins import str
+print(str('a')==bytes('a'))
+str('b')

After applying this, it prints out False.