[Refactor][Dataset] YesNo implementation

[krishnakalyan3]

Implementation for the YesNo dataset.

cc: @mthrok
Fixes: #910

The initial plan is to refactor the datasets (first part of the PR). Based on the complexity will create subsequent PRs.

[mthrok]

Looks good, thanks. Please address my few comments and also please fix the style check issue.

[mthrok]

@krishnakalyan3

The initial plan is to refactor the following datasets mentioned above in the first part of the PR.
Based on the complexity will create subsequent PRs.

Thanks, the code looks good. Since this PR is to aide the discussion, I will keep it as-is until we hear back on #910. (but because of the time of the year, I expect that we will not hear back very soon)

If you have extra time you would like to kill, then you can work on LJSPEECH too, but please keep it a separate PR. For CommonVoice currently #1126 is going on so you might want to hold off a bit.

[krishnakalyan3]

@mthrok wishing you a merry Christmas and a happy new year :).
Thanks for the feedback. I have made the necessary corrections.

[krishnakalyan3]

@mthrok I need your help again here. I am not sure how to resolve the style issue here. I am not exactly sure how to resolve this. This is what I was using.

[mthrok]

flake8 is no happy when indentation for continuous line is 4 spaces, because that is visually same as logical indentation. This often happens when if statement spans multiple lines or function indentation like this.

I think indenting 4 more spaces will resolve the issue.

def __init__(
        self,
        ...
) -> None:
    ...

[mthrok]

@mthrok wishing you a merry Christmas and a happy new year :).
Thanks for the feedback. I have made the necessary corrections.

@krishnakalyan3 thanks, wish you a happy holidays too :)