feature request: to_pynapple_tsgroup export Sorting properties (like SortingAnalyzer metadata)

[pas-calc]

The function to_pynapple_tsgroup (from spikeinterface.exporters) can export a BaseSorting or SortingAnalyzer to pynapple time series group (spike trains).

It is nice that it can also export metadata from SortingAnalyzer. However it currently does not support exporting properties from Sorting. For example when we read a sorting from Kilosort, the properites are automatically filled from the csv/tsv files.

For instance sorting.get_property("ch") will be the channel on which the neuron showed largest waveform.

For a sortinganalyzer run on that sorting+recording, we can get quality metrics. They will be similar to those precomputed by a sorter (like refractory period contamination, firing rate, ... )

For a sorting object, we could get_property_keys and then add the properties to the pynapple TsGroup metadata

[chrishalcrow]

Hello, that sounds good.

I think we'd just need to add a loop over the properties here:

spikeinterface/src/spikeinterface/exporters/to_pynapple.py

Line 81 in 456265c

and add them to the metadata_list as a dataframe. Sorting properties must be the length of the number of units, so should all be ok.

If you're willing to make a PR, I'll review it :)