477 n-dimensional tuning curves

[wulfdewolf]

This combines:

• compute_1d_tuning_curves
• compute_1d_tuning_curves_continuous
• compute_2d_tuning_curves
• compute_2d_tuning_curves_continuous

into a generalised n-dimensional compute_tuning_curves, which now returns an xarray.DataArray (new dependency).

The old functions are still available, but they call the new one under the hood and show a DeprecationWarning.
A couple of notes (these clarify why I had to make some changes to the old tuning curve function tests):

• for compute_1d_tuning_curves, we were using features.rate to scale the tuning curves to Hz, but if one passes epochs that are bigger than that of the input features, this results in an incorrect computation of the rate and thus incorrect tuning curves.
  FIXES: we now compute the rate as 1/np.mean(features.time_diff(epochs)).
  There further is an optional fs argument, for if you want to pass the exact sampling rate yourself.

• for compute_1d_tuning_curves_continuous and compute_2d_tuning_curves_continuous, we were using np.digitize to bin, but this does not include the rightmost edge , unlike np.histogram.
  FIXES: we now use np.histogram for all tuning curves.
  Hence, the new function is more consistent, as all tuning curves will now include the rightmost edge.

I have further updated all the notebooks and docs to use the generalised function.
The new xarray.DataArray output allows for some more concise plotting in a lot of the notebooks.

At the moment, the decoding is still separated into 1D and 2D, generalising that will be another PR.
For now, I am transforming the xarray.DataArrays into the respective old formats, such that the decoding functions still work.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

Files with missing lines	Coverage Δ
pynapple/core/interval_set.py	97.11% <ø> (ø)
pynapple/core/ts_group.py	92.88% <ø> (ø)
pynapple/process/__init__.py	100.00% <ø> (ø)
pynapple/process/tuning_curves.py	95.91% <100.00%> (-0.17%)	⬇️

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[sjvenditto]

I've left a few final comments that are more inconsequential, so let me know your thoughts on if you want to put in these small changes -- otherwise, this one's just about done!