Implement test suggestions, update expected values

[jacquesalice]

Running Client tests
  against Server: "sparc1.datalab.noirlab.edu"
  comparing to: tests.expected_pat   
  showact=True
  showcurl=False
  cls.client=(sparclclient:1.2.2b8, api:11.0, https://sparc1.datalab.noirlab.edu/sparc, client_
hash=, verbose=False, connect_timeout=1.1, read_timeout=5400.0)

All tests pass on sparc1 except for one (test_auth_retrieve_2), which is not working as it should. Logged in as the auth_user, it passes a total of 8 UUIDs (2 from each dataset -- DESI-EDR, BOSS-DR16, SDSS-DR16, and SDSS-DR17-test) to client.retrieve() without specifying any dataset. It should return 8 records, but it only returns 6 records and claims:

UserWarning: Some UUIDs were not found. 2 out of the 8 requested uuids have no records available in the SPARCL database associated with DataSets {'BOSS-DR16', 'DESI-EDR', 'SDSS-DR16'}.

Otherwise, this PR cleans up the other Auth tests and updates the expected values in expected_pat.py, which should help @pothiers with updating his expected values in expected_dev1.py.

[pothiers]

I thought we agreed you would work on factory-boy related stuff (in a separate file) and NOT client tests or expected values. I've been working there. Our changes will certainly conflict (and I've made a lot of changes). I don't expect to merge this.

[jacquesalice]

@pothiers No, I said I had already started working on implementing your suggested changes in the client tests and updating the expected values, and that I would start looking into factory-boy after. There is one client test that's still failing and you told me to wait until you've stabilized things before trying that one again. The reason I made this PR is because all of the other tests pass now (on pat), and I thought it would be helpful for you to see what was in expected_pat.py so you have an idea of what you need to include in your expected_dev1.py.

Plus, I thought I was in charge of adding+updating Auth-related client tests, as per DLS-505.

I don't think it's fair of you to Close this PR!

[jacquesalice]

Closing this PR because PR#36 has new changes to the same files.