Run pipelines on datasets

[jacob-rosenthal]

Implement a method to run a preprocessing pipeline on a dataset.
Should basically be a convenience function for running pipeline on each individual image.

Pseudocode:

mydataset = pathml.datasets.PESO.download()
mypipeline = pathml.preprocessing.default_pipeline()

mypipeline.run(mydataset)
### should be equivalent to:
for wsi in mydataset:
    mypipeline.run(wsi)

[jacob-rosenthal]

This should ideally be parallelizable

[jacob-rosenthal]

Should probably use something like this: https://joblib.readthedocs.io/en/latest/parallel.html#