Pynapple

- Morlet wavelets spectrogram with utility for plotting the wavelets.
- (Mean) Power spectral density. Returns a Pandas DataFrame.
- Convolve function works for any dimension of time series and any dimensions of kernel.
- `dtype` in count function
- `get_slice`: public method with a simplified API, argument start, end, time_units. returns a slice that matches behavior of Base.get.
- `_get_slice`: private method, adds the argument "mode" this can be: "after_t", "before_t", "closest_t", "restrict".
- `split` method for IntervalSet. Argument is `interval_size` in time unit.
- Changed os import to pathlib.
- Fixed pickling issue. TsGroup can now be saved as pickle.
- TsGroup can be created from an iterable of Ts/Tsd objects.
- IntervalSet can be created from (start, end) pairs

- Full lazy-loading for NWB file.
- Parameter `load_array` for time series can prevent loading zarr array
- Function to merge a list of `TsGroup`

- Full `pynajax` backend compatibility
- Fixed `TsdFrame` column slicing

- Fixing IntervalSet `__repr__`. Tabulate conflict with numpy 1.26.

- Improving `__repr__` for all objects.
- TsGroup `__getattr__` and `__setattr__` added to access metadata columns directly
- TsGroup `__setitem__` now allows changes directly to metadata
- TsGroup `__getitem__` returns column of metadata if passed as string

- `smooth` now takes standard deviation in time units
- Fixed `TsGroup` saving method.
- `__getattr__` of `BaseTsd` allow numpy functions to be attached as attributes of Tsd objects
- Added `get` method for `TsGroup`
- Tsds can be concatenate vertically if time indexes matches.