Conda Mac: Flaky test_hypothesis_mean_agg

[IvoDD]

Describe the bug

The test_hypothesis_mean_agg occasionally fails on the conda mac build as seen here.

FAILED tests/unit/arcticdb/version_store/test_aggregation.py::test_hypothesis_mean_agg[EncodingVersion.V2] - AssertionError: DataFrame.iloc[:, 0] (column name="a") are different

DataFrame.iloc[:, 0] (column name="a") values are different (100.0 %)
[index]: [0]
[left]:  [0.3333333333333333]
[right]: [0.0]
At positional index 0, first diff: 0.3333333333333333 != 0.0
Falsifying example: test_hypothesis_mean_agg(
    lmdb_version_store=NativeVersionStore: Library: test_hypothesis_mean_agg_Encod.54_2024-01-24T12_44_01_701[287](https://github.com/man-group/ArcticDB/actions/runs/7639977500/job/20814128990?pr=1265#step:9:288)_v2, Primary Storage: lmdb_storage.,
    df=
          grouping_column             a
        0               0  1.000000e+00
        1               0  9.007199e+15
        2               0 -9.007199e+15
    ,
)

You can reproduce this example by temporarily adding @reproduce_failure('6.72.4', b'AXicY2YAAkYmBkYQLf///wcgBeYxwnggDgBjlQYn') as a decorator on your test case

Steps/Code to Reproduce

Rerun test_hypothesis_mean_agg as in: https://github.com/man-group/ArcticDB/actions/runs/7639977500/job/20814128990?pr=1265

Expected Results

Test doesn't fail.

OS, Python Version and ArcticDB Version

See https://github.com/man-group/ArcticDB/actions/runs/7639977500/job/20814128990?pr=1265

Backend storage used

No response

Additional Context

No response