Add rejection sampling for LPM effect

This PR adds a rejection sampling for the LPM effect in discrete bremsstrahlung/photopairproduction losses. It also updates the proposal version to 7.5.0 (as this is the version where the LPM effect for photopairproduction has been implemented)

A comparison between LPM/non-LPM showers, and a cross-comparison with CORSIKA 7 is shown here

A detailed discussion about the theory and methodology has been given on these slides

I might still want to add a flexible way to enable/disable the LPM effect. Furthermore, I might want to add the LPM effect for electron-positron pair production (which is, as far as I know, only important for very-high-energy muons in dense media, as the cross sections itself is already negligible for electrons).

Note that this only adds the LPM effect for interactions, but not for continuous losses. However, this should be delegated to a different PR (and at least in the first comparison, the effect appears to be negligible for now).