Provide ability to reset index and time index on DataTable via method call

[thehomebrewnerd]

Currently if an index or time index is set on a DataTable they can be reset by calling dt.index = None or dt.time_index = None. However, this modifies the existing DataTable in place.

If we attempt to call dt.set_index(None) to reset the index, an error is raised. We should either update the existing set_index and set_time_index methods to allow a None value which will reset the index, or add new reset_index and reset_time_index methods.

To be most consistent with pandas, the approach should be to create new reset_index and reset_time_index methods on DataTable which return a new DataTable object with the index cleared.

New tests should also be added to test these methods.

[tyler3991]

we should implement this after #483. Ice boxing for now.

[thehomebrewnerd]

With the accessor approach calling df.ww.set_index(None) or df.ww.set_time_index(None) are valid calls and no longer result in an error.

@gsheni, @tamargrey Do you think we need to implement reset_index and reset_time_index methods, or can we close this issue?

[tamargrey]

If I were a user I'd probably go looking for a reset_index function before supplying the set_index function with None, but this probably isn't super pressing.

Another difference is also that the change only happens inplace now, and if we view this issue as there being a problem with
uneven behavior between the two index setters, then that problem no long exists.