QC for new DFE: "generic" human DFE

[chriscrsmith]

PR for new model:
#1505

(edited)

Original paper:
preprint: Kyriazis et al. 2022
ahead of print: Kyriazis et al. 2023

Parameter values:
Using values from "Description of simulation models" in supp material section 4:

• Gamma mean and shape: -0.0131 and 0.186
• Discretized h-s relationship: coefficients [0.45, 0.2, 0.05, 0] and breaks [0.001, 0.01, 0.1]
• proportion 0.003 lethal mutations (out of all deleterious mutations)

Potential issues:

• The proposed model calls for a deleterious mutation rate of 0.63 due to increased mutation rate in coding regions. We're not using rate maps in stdpopsim currently, so the analysis2 group decided to ignore this for now, but leave a note somewhere.