A critique of Boltzmann solution methods for calculating the longitudinal diffusion coefficient of electrons in gases