Workaround for nightly test failures (#1765)

After some debugging, the nightly tests fail when using parallel test
execution on seemingly arbitrary tests; but I can't replicate that
elsewhere. This PR works around the issue by disabling parallel test
execution and extending the time budget for nightly tests (since it does
indeed take longer when run serially).

---

prior, the commits in this PR were used for debugging

Author: mpharrigan

.github/workflows/nightly.yaml

@@ -30,7 +30,7 @@ permissions: read-all
 jobs:
   pytest:
     runs-on: ubuntu-22.04
-    timeout-minutes: 60
+    timeout-minutes: 120
     steps:
       - uses: actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 # v4
       - uses: ts-graphviz/setup-graphviz@c001ccfb5aff62e28bda6a6c39b59a7e061be5b9 # v1
@@ -43,4 +43,4 @@ jobs:
           pip install -r dev_tools/requirements/envs/pytest.env.txt
           pip install --no-deps -e .
       - run: |
-          check/pytest --durations=10
+          check/pytest --durations=10 -v -n=0