[Python][Docs] Update datasets user guide with more details on Partitioning(Factory)

[asfimport]

Currently, the datasets user guide has a section about the different partitioning flavors: https://arrow.apache.org/docs/python/dataset.html#different-partitioning-schemes. However, for the case where the partitioning schema is inferred from the file paths, this only shows the "short-cut" way to do this, and not the equivalent explicit method to create a factory object with ds.Hive/DirectoryPartitioning.discover(..) call, which allows to specify more options (eg using dictionary types or not, etc).

Reporter: Joris Van den Bossche / @jorisvandenbossche

_{Note: This issue was originally created as ARROW-13791. Please see the migration documentation for further details.}