fix: allow parsing of mutt aliases using non-ASCII characters

[romanofski]

This uses the improvements from purebred-mua/purebred-email#88 as the current code is unable to parse addresses in the mutt alias file using non-ASCII characters.

This uses the text address parser since the ByteString parser would expect the addresses in the mutt alias file to be content encoded. Don't think we want to expect users to do this.

Fixes #515

[frasertweedale]

faily test:

     exit vim                   (0.03s)
      user error (Text.Regex.Posix.ByteString died: (ReturnCode 8,"Unmatched ( or \\("))

[romanofski]

Yeah... I'll need to reserve some time to have a look.

[romanofski]

Issue is in tasty-tmux as it can't match unicode characters. See purebred-mua/tasty-tmux#20

[romanofski]

Yay Christmas present... Seems we don't need to jump to a different regex engine alltogether. For whatever reason I don't fully grok yet is that the ByteString with extended UTF8 characters was causing failures. Using Text.encodeUtf8 seems to properly encode the string.