ULTRa L1c write counts at a higher nside to combat aliasing

[lacoak21]

Apparently counts are required to be integral values at l2- this causes weird aliasing counts across pixels in l2 maps when we push counts. We have to push them to the map because they need to be integral values. To fix this, Nick proposed writing counts out at a higher nside and then downsampling them at l2. Im not 100% following this but I think it generally makes sense to me.

IF we are going from more multiple pixels to one pixels at l2 we will have less edge issues that are caused by these integral counts.

[lacoak21]

Nick said to add this to the l1c counts metadata in var notes

Yeah, I think you can just say something along the lines of: "Counts healpix maps are sampled at finer resolution to help maintain the pointing accuracy for each event.  Since count maps are a binned integral quantity, they necessarily require a non-spun approach per Pointing, unlike exposure time and sensitivities."