Unicode support

[kmod]

Hey @vinzenz, we're trying to add unicode support to Pyston and it looks like pypa doesn't support unicode literals -- ex it will treat u"\u0180" as the length-6 string "\\u0180". Is this something that could get added?

Right now we're using UTF8 to store the unicode data as part of the AST, but I don't mind switching that to something else (I think our unicode implementation uses UTF-32 anyway).

[undingen]

This got implemented in #20.
What AFAIK still missing for complete unicode support is specifying a source file encoding.