Sequential Shuffling BasicPersister

[pickles-bread-and-butter]

Overview:

We want to add shuffling to a non-spark dataset but we don't need to use Flyte/DynamoDB/etc to do this. We can simply sequentially shuffle locally and not worry about the runtime being O(N) for right now. We can come back later in time to update this to user multi-proc approach.

Acceptance Criteria:

• Sequential shuffling of local basic persistor can be turned on
• Tests exist to test that the sequential shuffling shuffles but persists all of the data correctly