theory covmat plotting depends on both point_prescriptions and point_prescription

[RoyStegeman]

Many (or maybe all) of the covmat plotting tools take theory_covmat_custom as input for the covmat. In #2198 this function was redefined to be the sum over the covmats generated for different point prescriptions in theory_covmat_custom_per_prescription, however, to give a label to the plot it assumes the covmat corresponds to a single point prescripton.

A simple solution would be to replace the input theory_covmat_custom with theory_covmat_custom_per_prescription for all the relevant functions. Alternatively, we may want to be able to plot the sum of the covmats as well, but that would require updating the labeling of the plots (perhaps by explicitly adding all point prescriptions that contribute to the plot?). We'll have to choose which option we want to go with.

P.S. even now the plotting functionality still works, it just requires to pass the same point prescription twice, once to point_prescription and once as a list with a single item to point_prescriptions, e.g.

point_prescriptions: ['7 point']
point_prescription: '7 point'

[scarlehoff]

I like the second option, plotting the sum and have plot_7points_3points_aThirdQ_BatmanSymbol.png