Tests failing with data_too_large health check

Begin with the failing cases in array_api_tests/test_operators_and_elementwise_functions.py and the linked GitHub Actions run; compare their Hypothesis settings with the passing local environment. Determine why data_too_large is triggered and verify the affected bitwise, division, comparison, and multiplication tests pass without an unexplained suppression.

See for instance https://github.com/data-apis/array-api-strict/actions/runs/9686873755/job/26730272068?pr=35

FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_bitwise_or[bitwise_or(x1, x2)] - hypothesis.errors.FailedHealthCheck: Examples routinely exceeded the max allowable size. (20 examples overran while generating 9 valid ones). Generating examples this large will usually lead to bad results. You could try setting max_size parameters on your collections and turning max_leaves down on recursive() calls.
See https://hypothesis.readthedocs.io/en/latest/healthchecks.html for more information about this. If you want to disable just this health check, add HealthCheck.data_too_large to the suppress_health_check settings for this test.
FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_bitwise_or[__or__(x, s)] - hypothesis.errors.FailedHealthCheck: Examples routinely exceeded the max allowable size. (20 examples overran while generating 8 valid ones). Generating examples this large will usually lead to bad results. You could try setting max_size parameters on your collections and turning max_leaves down on recursive() calls.
See https://hypothesis.readthedocs.io/en/latest/healthchecks.html for more information about this. If you want to disable just this health check, add HealthCheck.data_too_large to the suppress_health_check settings for this test.
FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_bitwise_xor[bitwise_xor(x1, x2)] - hypothesis.errors.FailedHealthCheck: Examples routinely exceeded the max allowable size. (20 examples overran while generating 9 valid ones). Generating examples this large will usually lead to bad results. You could try setting max_size parameters on your collections and turning max_leaves down on recursive() calls.
See https://hypothesis.readthedocs.io/en/latest/healthchecks.html for more information about this. If you want to disable just this health check, add HealthCheck.data_too_large to the suppress_health_check settings for this test.
FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_bitwise_xor[__ixor__(x1, x2)] - hypothesis.errors.FailedHealthCheck: Examples routinely exceeded the max allowable size. (20 examples overran while generating 9 valid ones). Generating examples this large will usually lead to bad results. You could try setting max_size parameters on your collections and turning max_leaves down on recursive() calls.
See https://hypothesis.readthedocs.io/en/latest/healthchecks.html for more information about this. If you want to disable just this health check, add HealthCheck.data_too_large to the suppress_health_check settings for this test.
FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_divide[__itruediv__(x1, x2)] - hypothesis.errors.FailedHealthCheck: Examples routinely exceeded the max allowable size. (20 examples overran while generating 9 valid ones). Generating examples this large will usually lead to bad results. You could try setting max_size parameters on your collections and turning max_leaves down on recursive() calls.
See https://hypothesis.readthedocs.io/en/latest/healthchecks.html for more information about this. If you want to disable just this health check, add HealthCheck.data_too_large to the suppress_health_check settings for this test.
FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_floor_divide[__floordiv__(x, s)] - hypothesis.errors.FailedHealthCheck: Examples routinely exceeded the max allowable size. (20 examples overran while generating 9 valid ones). Generating examples this large will usually lead to bad results. You could try setting max_size parameters on your collections and turning max_leaves down on recursive() calls.
See https://hypothesis.readthedocs.io/en/latest/healthchecks.html for more information about this. If you want to disable just this health check, add HealthCheck.data_too_large to the suppress_health_check settings for this test.
FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_greater[greater(x1, x2)] - hypothesis.errors.FailedHealthCheck: Examples routinely exceeded the max allowable size. (20 examples overran while generating 9 valid ones). Generating examples this large will usually lead to bad results. You could try setting max_size parameters on your collections and turning max_leaves down on recursive() calls.
See https://hypothesis.readthedocs.io/en/latest/healthchecks.html for more information about this. If you want to disable just this health check, add HealthCheck.data_too_large to the suppress_health_check settings for this test.
FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_less_equal[__le__(x1, x2)] - hypothesis.errors.FailedHealthCheck: Examples routinely exceeded the max allowable size. (20 examples overran while generating 9 valid ones). Generating examples this large will usually lead to bad results. You could try setting max_size parameters on your collections and turning max_leaves down on recursive() calls.
See https://hypothesis.readthedocs.io/en/latest/healthchecks.html for more information about this. If you want to disable just this health check, add HealthCheck.data_too_large to the suppress_health_check settings for this test.
FAILED array_api_tests/test_operators_and_elementwise_functions.py::test_multiply[__mul__(x1, x2)] - hypothesis.errors.FailedHealthCheck: Examples routinely exceeded the max allowable size. (20 examples overran while generating 9 valid ones). Generating examples this large will usually lead to bad results. You could try setting max_size parameters on your collections and turning max_leaves down on recursive() calls.
See https://hypothesis.readthedocs.io/en/latest/healthchecks.html for more information about this. If you want to disable just this health check, add HealthCheck.data_too_large to the suppress_health_check settings for this test.

I can't reproduce this locally, and I don't know what is causing this. The same tests on that PR were passing previously, so I'm assuming something changed in hypothesis.

For now, I am just going to suppress this health check, because I need to get that array-api-strict PR merged and released, but we should figure out what is going on.