Enhance estimate of astrophysical foreground rates

We should determine what the consensus population model for calculating p_astro is, and incorporate the full distribution alongside AMPLFI's posteriors. In addition to being more accurate, this would also give us error bars on the p_astro calculation, which might be valuable. 