Plotting grid of pre-fit and post-fit distributions relies on copy of cabinetry method

[ekauffma]

cabinetry's visualize.data_mc method returns matplotlib Figure objects, which are difficult to manipulate for display within the notebook. Ideally we want to display a grid of plots corresponding to pre and post-fit distributions for each channel in order to save space in the notebook. Currently there is a similar method that accomplishes this but copies too much code from cabinetry. This may be better-solved within cabinetry, see https://github.com/scikit-hep/cabinetry/issues/142