Add options to plotting script

[bryates]

Indeed, I would like to comment on this. In my opinion, we should have the option to display both systematic and statistical errors separately and in quadrature. If I understand correctly, the reason the statistical error isn't shown in the CRs is the assumption that the samples are large enough to make this error negligible. However, I'd prefer to validate this assumption and ensure that we can safely disregard MC statistical uncertainties.

Perhaps we could use the booleans unblind, plot_syst_err, and plot_stat_err independently of one another and avoid using pairs of them in the same if statement. This would give people full control and allow them to create combinations of boolean values depending on what they want to check.

Originally posted by @anpicci in #492 (comment)