Fix automatic contributor refresh

[MoralCode]

Description
in #3748 (comment) @shlokgilda pointed out that A) a bug was preventing the periodic hourly contributor profile refresh task from running, and b) this task was not pulling any useful updated info from the users profile

This PR fixes #3748

Notes for Reviewers
this PR by itself may be broken if #3754 is not also merged - they should be independent as far as merge order though

Signed commits

• Yes, I signed my commits.

[shlokgilda]

LGTM.

[MoralCode]

tested and now working

2026-03-04 22:11:59 d0c4fff081f1 process_contributors[283] INFO Contributor 177 of 477
2026-03-04 22:11:59 d0c4fff081f1 process_contributors[283] INFO Contributor 178 of 477
2026-03-04 22:11:59 d0c4fff081f1 process_contributors[283] INFO Contributor 179 of 477
2026-03-04 22:11:59 d0c4fff081f1 process_contributors[283] INFO Contributor 180 of 477
2026-03-04 22:12:00 d0c4fff081f1 process_contributors[283] INFO Contributor 181 of 477
2026-03-04 22:12:00 d0c4fff081f1 process_contributors[283] INFO Contributor 182 of 477
2026-03-04 22:12:00 d0c4fff081f1 process_contributors[283] INFO Contributor 183 of 477

[sgoggins]

This one seems to depend on the other one I just reviewed, since its importing it. Lets make sure we merge them both or merge neither of them.

[MoralCode]

theres a dependency in that sense but the two PRs are independent/can be merged in any order since the divide between them is inside vs outside that data access function

they dont have to both be merged, but thats what i would strongly recommend doing. the testing i did above was with all 4 prs

[MoralCode]

Pushed additional commits to partially address this failure we are seeing in prod

Stack Trace

Traceback (most recent call last):
  File "/augur/.venv/lib/python3.11/site-packages/celery/app/trace.py", line 453, in trace_task
    R = retval = fun(*args, **kwargs)
                 ^^^^^^^^^^^^^^^^^^^^
  File "/augur/.venv/lib/python3.11/site-packages/celery/app/trace.py", line 736, in __protected_call__
    return self.run(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/augur/augur/tasks/github/contributors.py", line 52, in process_contributors
    data = retrieve_dict_data(url, key_auth, logger)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/augur/augur/tasks/github/contributors.py", line 84, in retrieve_dict_data
    page_data = response.json()
                ^^^^^^^^^^^^^^^
  File "/augur/.venv/lib/python3.11/site-packages/httpx/_models.py", line 743, in json
    return jsonlib.loads(self.text, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/json/__init__.py", line 346, in loads
    return _default_decoder.decode(s)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/json/decoder.py", line 337, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
               ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/usr/local/lib/python3.11/json/decoder.py", line 355, in raw_decode
    raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

TL;DR it seems like one contributor is returning bad data in a way that raises an exception and isnt caught/skipped by existing logic.

I have manually checked a few users to see what the issue could be (my own, the GH ghost user, and a user i know has a private profile) and none returned non-JSON data.

The commits i added allow the process to skip and log contributors that demonstrate this error so the task can proceed.

Waiting for an oppotunity to redeploy this to our prod and test