The stream names of neuralynx are hard to use and affected by numpy > 2.1 changes

[h-mayorquin]

I’m looking at the testing data in the gin repo ("neuralynx" / "Cheetah_v6.4.1dev" / "original_data"), and the stream names are as follows:

I’m not very familiar with this reader, so I’m not sure whether these names can be improved to something more meaningful.

Why is this a problem? Users rely on the stream name to fetch information, and these names are hard to type and remember. More importantly, with recent changes in how NumPy represents scalar values, some previously working code is now broken.