Rework neutrino simulation pipeline & philosophy

[eduard322]

Currently, neutrinos are simulated in not straightforward and correct way (see https://indico.cern.ch/event/1594362/contributions/6838164/attachments/3193049/5683152/genie_pythia.pdf). Let's try to move to gevgen_fnal -> FairShip approach.

Issues that we currently have:

1. scattering only on one nucleus.
2. Pt cut for P > 50 GeV
3. Calculation of neutrino yield: fixed cross-section and energy (should be changed to only $\sigma(E)$).

gevgen_fnal fixes that.