Handle characters with different encodings

[leafo]

filenames with characters that aren't utf-8 encoded will cause the extraction to fail. We should be able to automatically convert between the encoding to avoid the error.

https://twitter.com/Tryall_YT/status/891304284284997632

Go throws the error: 'ERROR: invalid byte sequence for encoding "UTF8"

[fasterthanlime]

Sooooooooo

Paths encoding in .zip are a bit of a mess (in practice) (/cc @GranPC)

In theory, zip has a flag that you can set that says "all the paths are UTF-8" - if it's not set, then... they can be anything.

We've had a similar problem with butler's unzip command (for the app) and Jesus is working on a method that autodetects the encoding of paths if the utf-8 flag is not set. It hasn't landed yet, but when it does, we should switch zipserver over to https://github.com/itchio/arkive, which will solve this issue.