Create script to consolidate data preparation pipeline

[rchang0501]

Currently, we have multiple independent scripts + notebooks that perform our data preparation.

The goal of this task is to consolidate all these scripts into one big script that we can run for data preparation down the line.

The consolidated script should be able to take in parameters such as time step and fine area division. Based on this information, it should output a new training set with those time and space parameters.