Provide utility functions to build contact models

[janosg]

What would you like to enhance and why? Is it related to an issue/problem?

In our estimation repo we use quite many different ways to work with contact models. Some of them are slow. I suggest to implement a few fast functions that can be used instead.

Describe the solution you'd like

def reduce_contacts(contacts, multiplier, contact_type, states=None, condition=None):
    Args:
        contacts (pandas.Series or numpy.ndarray)
        multiplier (float)
        contact_type (str): One of "recurrent", "non-recurrent"
        states (pandas.DataFrame, optional): Only needed if condition is not None
        condition (str): eval string that returns boolean Series. 

To Do

• add more relevant interfaces