angle-dependence of photocathode coverage

[JostMigenda]

Currently, the photocathode coverage is defined as 40%, regardless of angle. In reality, this depends on the angle of incidence of each photon (see fig 4.5 in this thesis).

According to Michael Smy, Super-K uses a MC simulation to determine the precise angle-dependence. Since this is dependent on the detector design, it might not be worth the effort to do a full MC simulation until the design is finalized, but maybe we can use some approximation in the meantime?

[JostMigenda]

This commit adds basic dependence on theta. The change in reconstructed energies is very small, so this should be sufficient until the HK design is finalized.