Fixing encoding errors in FacebookMessenger parser

[UntriexTv]

Fix for #73 working on my side.
The reason for invalid encoding was because of non ASCII characters being encoded twice.
Explanation
The code now works without problems and gets rid of all emojis and wrongly encoded characters.
The slovak characters mentioned in the issue are fixed now:

[joweich]

@UntriexTv Makes sense! I'm able to connect some dots now: We had a related issue for the Instagram Parser in #65 (see here).
It got resolved by @BlueishTint by going with

  "author": mess["sender_name"].encode("latin-1").decode("utf-8"),
  "message": body.encode("latin-1").decode("utf-8")

when composing the parsed message into the dict to return.
And this is the same solution like the first proposed resolution in the stackoverflow thread you linked.
Did you try this approach as well?
I think it's cleaner and has probably better performance since regex matching is quite expensive.

[UntriexTv]

EDIT:
Ah yes sorry, I got it wrong, I'll implement this same fix for the Facebook parser. My bad

[UntriexTv]

@joweich Tried with this fix too. Seems to work perfectly

[joweich]

@UntriexTv Thank you for your efforts and welcome as a contributor to this project! 🤗