DM-50987: Add sattle to pipelines

[bsmartradio]

No description provided.

[kfindeisen]

The most glaring issue is the code duplication with lsst/ip_diffim#418, but I also have one big-picture question: what, exactly, is this code for? It doesn't seem to use the Sattle results for anything. Can it all just be removed?

[ebellm]

@kfindeisen The big picture is that sattle needs the exact exposure time and the boresight ra & dec to populate a cache that will later be used in ip_diffim.detectAndMeasure. That cache takes a small amount of time to populate, so we want to call it early in the pipeline execution process so we're not waiting on it later during detectAndMeasure. This call (to /visit_cache) doesn't return any data to pipelines.

(We have taken steps on the service side to make the service idempotent and able to ignore 189 duplicate requests.)

[kfindeisen]

The big picture is that sattle needs the exact exposure time and the boresight ra & dec to populate a cache that will later be used in ip_diffim.detectAndMeasure. That cache takes a small amount of time to populate, so we want to call it early in the pipeline execution process so we're not waiting on it later during detectAndMeasure.

I think a comment to that effect would be helpful (though, again, factoring the code might make that unnecessary if the new code is self-descriptive enough).

[kfindeisen]

Looks good. As before, many of my comments from lsst/ip_diffim#418 apply here as well.