Reconfigure spike time processing module creation to match NWB best practice

[tmchartrand]

Not a major issue, but should be an easy fix.
The way that IPFX creates and adds the spike time ProcessingModule to save spike time info apparently doesn't match the intended usage of the NWB standard. See the example here: https://pynwb.readthedocs.io/en/stable/tutorials/general/file.html#behavior-processing-module
This leads to a generic name ("ProcessingModule") when the DANDI metadata extraction tries to look up variables measured, as opposed to a more specific name as intended, e.g. "spikes". See: https://dandiarchive.org/dandiset/000288