Add sorting reader

[JoeZiminski]

This is a rough prototyping PR to add spikes loaded from kilosort output to the GUI. It is branched from #22. Currently it does not load spikeinterface sorting analyzer, that is reserved for a separate PR.

Since #22 makes it possible to show only a single shank, there are four cases to consider:

1. The file is displayed with all shanks, and the sorting contains all shanks
2. The file is displayed with all shanks, but the sorting is for a single shank
3. The file displays a single shank, and the sorting contains all shanks
4. The file displays a single shank, and the sorting is for a single shank

Currently the user must input the shank that the sorting is one (None for all) because I couldn't quickly figure out a way to infer the shank from the sorting output, however it is probably possible.

TODO MAJOR:

• it is also possible that the raw data is split per shank. This case is not currently handled as it assumes the data is always multi-shank (even if we are only viewing a single shank).

TODO:

• refactor
• get feedback
• check carefully and think about all possible use cases, for example I have not tested the 1-shank case yet
• write some tests

An example script:

from viewephys.gui import EphysBinViewer
from pathlib import Path
import easyqc
import numpy as np

app = easyqc.qt.create_app()

sorting_path = r"Y:\...\path_to_sorter_output"


# i.e. only display the first shank, + the sorting is multi-shank
# (the software will figure out displaying the correct spikes)
viewer = EphysBinViewer(
    r"path/to/data/g0_tcat.imec0.ap.bin",
    shank_idx=1, sorting_path=dat_path, sorting_shank_idx=None,
)

app.exec()