Fit SED without any photometric bands

[Chamani8]

The F19 sample of stars in optical are better fit without any photometric bands. However, measure_extinction uses photometric bands in many different places, that simply commenting out the photometric bands in the .dat files produces an error. The current solution is to use only one band that interferes with the fitting minimally, e.g. the J band.