Fix racing in shared memory creation

[XiaohanZhangCMU]

Description of changes:

This PR hopefully addresses several issues raised in dataset creation for certain OS. The theory is that certain OS is slow in propagating shared memory across processes, so the non-local leaders need more wait time on the local leader.

Issue #, if available:

• Related Cannot Load MDS Dataset #824
• Related Race Condition in base.shared.get_shm_prefix (possible general SharedMemory issue) #901
• Related StreamingDataset gives FileExistsError when called with multiprocessing #884
• Related File exists: '/000000_epoch_shape' when using the ddp strategy from pytorch lightning #767
• Related 'File exists: "/00000_locals"' when integrated with deepspeed training scripts #717

Merge Checklist:

Put an x without space in the boxes that apply. If you are unsure about any checklist, please don't hesitate to ask. We are here to help! This is simply a reminder of what we are going to look for before merging your pull request.

General

• I have read the contributor guidelines
• This is a documentation change or typo fix. If so, skip the rest of this checklist.
• I certify that the changes I am introducing will be backward compatible, and I have discussed concerns about this, if any, with the MosaicML team.
• I have updated any necessary documentation, including README and API docs (if appropriate).

Tests

• I ran pre-commit on my change. (check out the pre-commit section of prerequisites)
• I have added tests that prove my fix is effective or that my feature works (if appropriate).
• I ran the tests locally to make sure it pass. (check out testing)
• I have added unit and/or integration tests as appropriate to ensure backward compatibility of the changes.