[Feature Request] Time series processing directly on dataframes

[samgdotson]

Currently, all functions in tsprocess.py accept a path argument, which PyGenesys attempts to read into a pandas dataframe. These functions do NOT accept a dataframe itself -- which should change.

This issue can be closed when the following features have been added to the time series processing tools:

• The main argument should accept a string OR a dataframe.
• A "column" parameter to specify which columns in the dataframe should be used (this is also helpful for reading in dataframes with more than one column!) However, PyGenesys will still assume that the index column is called "time."