Improve halo angular momentum diffusion model in nodeOperatorHaloAngularMomentumVitvitska2002

[abensonca]

Revision 96c2959 introduced into the nodeOperatorHaloAngularMomentumVitvitska2002 class a simple model for the diffusion in angular momentum of halos caused by subresolution accretion. Previously, we accounted only for the mean angular momentum of this subresolution accretion, but this is typically very close to zero as accretion is close to isotropic. Nevertheless, the halo angular momentum should diffuse away from zero due to the random fluctuations in the angular momenta of the accreted (subresolution) halos.

In revision 96c2959 this diffusion is treated as a Weiner process with fixed variance per unit growth in $J = M_\mathrm{vir} V_\mathrm{vir} R_\mathrm{vir}$. In principle, it seems that we should be able to do a better calculation here - estimating the rms angular momentum per subresolution halo accreted (based on the orbital parameter distribution), estimating the variance due to the total number of subresolution halos accreted (e.g. assume them to be uncorrelated infalls, so simply sum the variances), and then sample from this (assuming the central limit theorem, so a normal distribution).