MNT/FIX: Update Lo exposure time calculations for off-angle bins

[greglucas]

Change Summary

This updates the exposure time calculation for Lo pointing sets.

• Accounts for off-angle bins by sampling a nominal (5) set of spins during a pointing at 4096 (nticks/spin?) samples, seeing what bins those fall into. We then histogram the off-angle components to get an off-angle distribution that can be applied evenly to all spin-bins.
• Then, the total pointing time is divided equally into all spin_bins, then multiplied by the off-angle fraction above.
• Finally, the goodtimes bins are turned into a fractional mask. The fractional mask is applied to (esa_energy_step, spin_bin), to account for how much time a given energy step and spin bin were a part of the total exposure time.

[subagonsouth]

LGTM