Ensure shards cannot be larger than 4GB

[snarayan21]

Description of changes:

Based on issue #659, users have seen that shards exceeding 4GB are not possible since we are encoding sample byte offsets using uint32, which has a max value of 2**32 - 1. To prevent users from hitting sample access errors much later, we instead will throw an error here.

If there are future use cases that MUST use shards larger than 4GB, we can revisit this and increase the byte offset encoding type to uint64 if applicable. This is currently not the case for any workload we support, including vision.

Added unit tests to all 3 writer types ensuring that the error is thrown correctly.

Issue #, if available:

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.

[karan6181]

LGTM. Thank You!