Add unique identifier to data

[TTitcombe]

Feature Description

Update the data partitioning function to apply a unique ID to each datapoint (shared across datasets).
The simplest solution would be an integer ID, however (fake) emails or names would also be acceptable to simulate data subjects

Is your feature request related to a problem?

We need some way of linking datapoints across datasets

Additional Context

Part of EPIC #2
Dependent on #4