Building a Sarracen DF is undocumented

[JBorrow]

At present, it is not clear how to build a sarracen dataframe from e.g. a list of particle positions and smoothing lengths. I think this should be documented somewhere for people who do not want to load using pandas.

[JBorrow]

openjournals/joss-reviews#5263

[ttricco]

@JBorrow If I am understanding your issue correctly, it is to add documentation on how to create a SarracenDataFrame without requiring reading the data from file. Our assumption has been that read_csv() would be leveraged for general reading of SPH data, but we can add documentation on how to create a SarracenDataFrame without requiring file reading.

[JBorrow]

Yes, I think it would also help to generally document the 'required' elements of the DF for visualisation, as this can help people who are developing on top of sarracen.

This is useful if e.g. you want to do isolated tests or visualise some initial conditions just after you finished creating them (or even, say, if you want to fill a SDF and use it to write out ICs for PHANTOM or some other code...)