Logistic Regression prediction probabilities

Currently sklearn logistic classifier only supports predictions.

https://github.com/neilsummers/mleap/blob/master/python/mleap/sklearn/logistic.py
```
def serialize_to_bundle(self, transformer, path, model_name):

# compile tuples of model attributes to serialize
attributes = list()
attributes.append(('intercept', transformer.intercept_.tolist()[0]))
attributes.append(('coefficients', transformer.coef_.tolist()[0]))
attributes.append(('num_classes', 2)) # TODO: get number of classes from the transformer

# define node inputs and outputs
inputs = [{
"name": transformer.input_features,
"port": "features"
}]

outputs = [{
"name": transformer.prediction_column,
"port": "prediction"
}]

self.serialize(transformer, path, model_name, attributes, inputs, outputs)
```

Want to be able to access the probabilities from the classifier too.
This is already supported in the MLeap logistic model. Just need to add the appropriate outputs. I will create a PR with the change we made to access this for our own production model.