fixed a bug by changing erroneous R1r to correct R2r in JWL_f

[hackljf]

the JWL_f function in exactpack/solvers/riemann/utils.py evaluated part of the JWL $p(v,e)$ EOS as A * (1. - G / R1r) * exp(- R1r) + B * (1. - G / R1r) * exp(- R2r)

This should have read

A * (1. - G / R1r) * exp(- R1r) + B * (1. - G / R2r) * exp(- R2r)