Ionized electron momentum using Tomassini's model

[EyaDammak]

This pull request follows the PR #1195.
2 errors have been fixed:

• error in the second prefactor (laser_dp_second_prefactor) :
  $\alpha = 2n*-1 = 2Z \sqrt{(U_H/U_{ion})} - 1$
• Model not physically correct for our case
  The model for the transverse momentum spread $\sigma_{u_x}$ in linear polarization was taken from equation [14] of Schroeder article which results from an average on the transverse plan (what he calls "effects of off-axis ionisation of electrons") and only works for a gaussian laser model.
  This new PR implemented the result from Tomassini's article ( #https://doi.org/10.1017/hpl.2021.56, equation (12)), including the terms of order $\Delta^3$ and $\Delta^5$ for the rms electron momentum ($\sigma_{u_x}$ also called width_p in the src code):
  $\sigma_{u_x} = a_0 \Delta ( 1 + s1 * \Delta^2 + s2 * \Delta^4)$
  with $s1 = -7/4 + \alpha / 2$ and $s2 = (8\alpha^2 - 68\alpha+131)/16$

[MaxThevenet]

Thanks for this PR! Do we expect, from theory, that Tomassini's expression should always be more accurate than what is in Massimo's? Probably yes, but if not we could have both implemented and available. Let's have a quick chat.