Add algorithm types for the iterative learning primitive

[d8w]

The IterativeLearning primitive does not have proper algorithm types in the metadata. Algorithm type describes the underlying implementation of the primitive. Currently it is set to binary classification, which is only one of many possible applications. We need to tag this primitive with all applicable applications to make it easy for systems to automatically discover it.

What needs to be done?

1. We need to find suitable algorithm types defined in the D3M schema or add new ones there.
2. We also need to update the algorithm_types field in iterable_label.py.

[d8w]

Added a "iterative labeling" type to the schema. See MR https://gitlab.com/datadrivendiscovery/d3m/merge_requests/276