TypeError: must be str, not bytes

[TFWol]

Related to #27
It looks like if you run Lyndor on a Win10 machine with normal Locale settings, it will trip on the lines with .encode('utf-8') at the end with the error TypeError: must be str, not bytes.

I'm seeing the error at

Lyndor/save.py

Line 177 in 770483b

content_md.writelines('\n\n## ' + (chapters[chapter_count].text).encode('utf-8') + '\n')

I think it may also happen at similar lines if it were to reach those.
Maybe we need the original code as a try then have it use the .encode upon a Unicode error since the current change doesn't work for all use cases like hoped.

That or do a test branch that only uses Unicode to see if that works for all cases.

[ankitsejwal]

@thefinalwolf I've made the changes please have a look when you get time to see if this patches the issue.

[TFWol]

Yeah, that fixed this issue, thanks.