Send data mask from theory calculator into plotter

[pkienzle]

From @butlerpd

So we got this email on the help desk about the script for fitting data to the guinier_porod model with bumps was not working because bumps is not honoring the qmin (but does honor qmax). James Crake-Merani (ISIS intern) tried it and agrees with me that it should work but says it does not. He also tried doing the mask differently per Wojciech’s suggestion which also did not work.

This PR lets me fit a dataset with qmin and qmax specified using bumps.

To test, uncomment the following:

sasmodels/example/simul_fit.py

Line 62 in 973518c

#datasets[1].qmin, datasets[1].qmax = 1e-4, 4e-3

and run it in bumps:

bumps example/simul_fit.py -s

To test 2D data masks, uncomment the following:

sasmodels/example/fit.py

Line 17 in 2662896

#radial_data.qmin, radial_data.qmax = 0.01, 0.02

and run it in bumps:

bumps example/fit.py -s