Fix numpy failure case in det_cal

[mmccrackan]

Some obs_ids are silently failing on the LAT with the following error:

obs_1782017305_lati3_111: "Traceback (most recent call last):\n  File \"/usr/local/lib/python3.12/site-packages/sotodlib/site_pipeline/update_det_cal.py\"\
  , line 629, in get_cal_resset\n    if not ridx:  # Channel doesn't exist in IV analysis\n\
  \           ^^^^\nValueError: The truth value of an empty array is ambiguous. Use\
  \ `array.size > 0` to check that an array is not empty.\n"

This just adds a fix so that it will continue and not fail as was done for update_smurf_caldbs in #1619.

[ykyohei]

I tested update_det_cal for this obsevation at tiger3 and it didn't have any error and I'm bit confused.
Is it possible that this is due to race conditions of site pipeline..?

[mmccrackan]

I tested update_det_cal for this obsevation at tiger3 and it didn't have any error and I'm bit confused. Is it possible that this is due to race conditions of site pipeline..?

Multiple update_det_cal flows aren't permitted to run at the same time nor can they run at the same time as update_obsdb or update_smurf_caldbs so I wouldn't have thought it was a race condition.

I see many more of these kinds of failures in the failed_obsid.yaml file as well.

[ykyohei]

Okay I learned that this used to be just DeprecationWarning but from numpy 2.2.0 this leads to ValueError.
Thanks for flagging this!
https://numpy.org/devdocs/release/2.2.0-notes.html#expired-deprecations

[mmccrackan]

Okay I learned that this used to be just DeprecationWarning but from numpy 2.2.0 this leads to ValueError. Thanks for flagging this! https://numpy.org/devdocs/release/2.2.0-notes.html#expired-deprecations

Great, I am going to remove the obs_ids with this error from the failed_obs_ids file so they will re-run.