python 3.11 CI test failure with AzureBlobStore

[rkingsbury]

Since adding python 3.11 support (See #737 ), tests.stores.azure.test_multi_update frequently (but not always) fails using python 3.11. See this example workflow run.

I don't know the exact cause, but the issue is that the test condition is based on the execution time, with the assumption that a multiprocessed write will be faster than a single write:

        start = time.time()
        blobstore_multi.update(data, key=["task_id"])
        end = time.time()
        time_multi = end - start
    
        start = time.time()
        blobstore_two_docs.update(data, key=["task_id"])
        end = time.time()
        time_single = end - start
>       assert time_single > time_multi * (blobstore_multi.workers - 1) / (
            blobstore_two_docs.workers
        )
E       assert 0.26502466201782227 > ((0.5272753238677979 * (4 - 1)) / 4)
E        +  where 4 = <maggma.stores.azure.AzureBlobStore object at 0x7f0c5daa4810>.workers
E        +  and   4 = <maggma.stores.azure.AzureBlobStore object at 0x7f0c5daa4810>.workers

For whatever reason, in python 3.11 it seems that at least sometimes the single write is faster. I'm not sure whether this is due to a change from python 3.10 to 3.11, or some issue with GitHub python 3.11 workers, but regardless, I wonder @gpetretto if there is a different way you could write the assert statements that would test the functionality without being dependent on performance?