experimental DataFrame support

[jhgjeraker]

Overview

Experimental function to convert a list of events to DataFrame.
Requires the installation of additional packages.

pip install disruptive[extra]

The pandas package is not, and will not, be a dependency of
the core disruptive package. However, supporting DataFrames is a
significant enough convenience that we're experimenting with it here.
May be removed in the future if we decide that this is not a good idea.

The columns device_id, event_id, and event_type are static, then
one additional column is concatenated for every eventData field.

---

Changes to EventHistory.list_events()

Before:
EventHistory.list_events() output was list[Event].

Now:
EventHistory.list_events() output is EventHistory, the instance of a class that itself inherits the list type. This allows us to add functionality like to_dataframe() while preserving all the previous behavior expected from a list.

For an list_events() output events, the check

isinstance(events, list)

still holds and should not affect previous implementations.