Create tests for PR fix bug when logcat value which cannot be decode to utf-8

[Draqun]

Create tests for fix from link below.

#856

[Draqun]

Hi. Sadly I don't have enough skills to write tests. I don't know docker (ugly solution) and there is no requirements.txt file to build venv.
Sorry for that.

[AndreMiras]

No worries, this could be tricky.
I'm actually not sure how this error you fixed happens.
Anyway for testing it I would say it should involve mocking https://docs.python.org/3/library/unittest.mock.html
The way I would have imagined it is calling Buildozer.cmd() within a mocked Popen to make sure it .stdout.read() returns a non utf-8 decodable string.
So the patching could look like this (untested):

command = 'mycommand'
kwargs = { 'show_output': True }
with mock.patch('buildozer.Popen') as m_popen:
    # the command output could be were we want to play with unicode decode errors
    m_popen().stdout.read.side_effect = [b'mycommand output']
    buildozer.cmd(command, **kwargs)

That just to give the idea.

[AndreMiras]

I just gave it a try see #905
So yes definitely not the best use case to start playing with mocks