Hardwired free-free limits in kpkt()

[jhmatthews]

Similarly to the recent problem with Helium, where the gamma estimators were hardwired to use a BBody above 50eV, we have a similar thing in kpkt(). When we decide to create a free-free photon from a k->r channel, we enter the following statement:

  else if (destruction_choice <
       (mplasma->cooling_bftot +
        mplasma->cooling_bbtot + mplasma->cooling_ff))
    {               //this is a ff destruction

      /* The limits for ff emission are hard-wired: 40 microns ->
         twice energy of He II edge. Shouldn't be a problem unless
         we're in plasma with temperatures that gives significant ff
         emission outside this range. */

      *escape = 1;      //we are making an r-packet not exciting a macro atom

      *nres = -2;

      p->freq = one_ff (one, 7.5e12, 2.626e16); //get frequency of resulting energy packet

      return (0);
    }

where the frequency is chosen to be <2.626e16 and >7.5e12. I reckon these should probably be set to the edge of the frequency domain in the simulation, so one can in theory generate very blue or red photons as long as that is globally allowed in the simulation.

[jhmatthews]

Closed by commit agnwinds/python@3d5dfb6 -- when we call one_ff in kpkt now we either use band limits (ion cycles) or em_range.fmin and fmax (spectral cycles). Means we generate ff photons over full range in matom mode.