Correct reddened model band fluxes

[karllark]

The current reddened model band fluxes are computed by applying the extinction to the unreddened band fluxes at the effective/pivot wavelength. This is an approximation and is done to speed up the fitting.

The fully correct reddened band fluxes should be computed by applying the extinction to the model spectra and then integrating over the band response functions. This requires a model spectrum that covers the entire band response function, yet the model spectra used in the fitting are only those where there are observed spectra in the fitting. What is needed is to have a fairly low resolution spectrum over the full wavelength range of possible photometric bands that is used for this calculation.

What is needed:

1. Saving of the appropriate resolution spectrum over the full possible photometric wavelength range (e.g., 912 A to 32 micron at least).
2. Reading in these spectra into the fit model object
3. Reading in the band response functions for the observed bands
4. Code to apply the extinction to these spectra and integrate over the appropriate band response functions