Pin tornado<6.2 in nightly conda recipes

[pentschev]

#6668 missed pinning tornado on conda recipes, nightly builds still pick tornado 6.2. This PR adds the missing pin.

• Tests added / passed
• Passes pre-commit run --all-files

[pentschev]

cc @fjetter @graingert @wence-

[github-actions]

Unit Test Results

See test report for an extended history of previous test failures. This is useful for diagnosing flaky tests.

       15 files  ±    0         15 suites  ±0   6h 32m 56s ⏱️ + 2m 44s
  2 917 tests ±    0    2 833 ✔️ +    6    79 💤 ±  0  3 ❌  - 8  2 🔥 +2 
21 600 runs  +827  20 660 ✔️ +796  933 💤 +36  5 ❌  - 7  2 🔥 +2 

For more details on these failures and errors, see this check.

Results for commit b22a855. ± Comparison against base commit 75f4635.

[mrocklin]

I'm happy to press the green button here. I'm seeing some CI failures. Can someone (@pentschev ? ) go through and verify that these are unrelated? Are we still raising issues for CI failures? If so, maybe we should do that too? I apologize for being out of the loop here.

[pentschev]

MacOS errors are apparently being handled by #6671:

distributed/tests/test_asyncprocess.py::test_signal
distributed/tests/test_asyncprocess.py::test_terminate

Another one is #6582 , and the last one is distributed/deploy/tests/test_local.py::test_localcluster_start_exception, which I found no mentions elsewhere, so maybe a new flaky test?

In any case, this PR is only modifying the pin to the conda package. I'm not quite sure, but I think the conda package isn't built and tested in CI, or is it? If it is not, then I don't think any errors could be coming from this change.

[jrbourbeau]

Thanks @pentschev, this LGTM

I'm not quite sure, but I think the conda package isn't built and tested in CI, or is it? If it is not, then I don't think any errors could be coming from this change.

When files related to the conda package are changed we will build the package in CI. For example, that happened in this CI build for this PR. But we don't use the built package when running our test suite. So I agree any test failures here are unrelated to the changes in this PR.

That said, would you mind opening up issues for any flaky tests that popped up here and don't already have an issue?