adapt read_spikegadget to NPX2 Bennu headstage

[RobertoDF]

It doesnt fulfill all the plans outlined out in #418 but should work with NPX2 already.

[h-mayorquin]

Hi @RobertoDF, thanks for taking a swing at NP2 support. Currently this breaks the NP1 path: the new loop reads probeColumn from every SpikeChannel unconditionally, but NP1.0 SpikeChannel elements don't carry that attribute, so the existing NP1 test fails with KeyError: 'probeColumn'.

I've opened #441 with an alternative implementation that I think solves the same problem while keeping the NP1.0 path working. I also added a test dataset that we have for 2.0 four-shank. For 2.0 single-shank though, I am just inferring from the Trodes source code. It would be particularly valuable to have test data for that case. Is that the configuration you are using by any chance, and could you share data with us?

In any case, could you test that to see if it works your problem? I am happy to add you as a contributor in that PR for the records.

[RobertoDF]

amazing, ill close this