Internal data representation is relevant for Workflows blocks creators - this is the type that will be provided
by Execution Engine in runtime to the block that consumes input of this kind.
This kind represent batch of prediction metadata providing information about the context of prediction.
For example - whenever there is a workflow with multiple models - such that first model detect objects
and then other models make their predictions based on crops from first model detections - parent_id
helps to figure out which detection of the first model is associated to which downstream predictions.