Assist with broadcasting "per-group scalars" to per-det values based on det_info

[mhasself]

It might be handy to be able to use det_info to broadcast values indexed by det_info fields e.g. (wafer, band) into per-det values.

Suppose there's a LabelAxis with a compound name based on fields in det_info... and compound values:

>>> aman['dets:wafer_slot,dets:passband']
LabelAxis(4:'ws0,f090','ws0,f150','ws1,f090','ws1,f150')

Then you can broadcast a thing mapped to that axis

>>> aman._assignments['wafer_avg_solid_angle']
['dets:wafer_slot,dets:passband']

>>> core.axisman.broadcast(aman.wafer_avg_solid_angle, det_info=aman.det_info)
np.array([12., 12,. 12., ...])

Assume that last thing is shape ('dets',), because it resolved it all through the det_info...

(This sort of broadcasting is already done for metadata on load; e.g. abscal is stored by (wafer, band) but on load converted to per-detector values (duplicated) based on det_info.)