Modify dataframe_to_mds to accept streaming DF

[maddiedawson]

Description of changes:

Modify util to accept a streaming dataframe. Performs merge index per-batch.

Tested in https://e2-dogfood.staging.cloud.databricks.com/?o=6051921418418893#notebook/4230899825197178

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.

[xiaohanzhan-db]

Not sure why only outputting to local is possible here? What is the challenge to support original out (meaning local, remote or (local, remote))?

[maddiedawson]

We want to disallow concurrent writes to the top-level index.json. We need a lock file that is shared among all workers. This is implemented here using open with flags to ensure that opening the lock file fails if it already exists.