[FEA] Cosine similarity for HDBSCAN

[nivibilla]

Is your feature request related to a problem? Please describe.
I would like to use cosine similarity as a metric for HDBSCAN for high dimensional data without dimension reduction as its quite slow to do for large datasets

Describe the solution you'd like
Allow clusterer = cuml.cluster.hdbscan.HDBSCAN(min_cluster_size=50, metric='cosine', prediction_data=True)

Describe alternatives you've considered
normal DBSCAN supports this but this has issues with clusters of various sizes.

[nivibilla]

Trying this currently raises this error

cuml/python/cuml/cuml/cluster/hdbscan/hdbscan.pyx

Line 838 in fb4c8af

raise ValueError(f"metric '{self.metric}' not supported, only "

[divyegala]

@nivibilla you could l2-normalize your dataset for now and pass metric=euclidean for equivalent results to cosine distance.