Enhancement: Ensure compatibility with latest scikit-learn releases

[AnnaBobasheva]

Hi tslearn team,

I'm encountering warnings when callin fit_transform and fit_predict methods with recent versions of scikit-learn v1.6.1
It appears that tslearn relies on internal APIs or behaviors that have changed in newer versions of scikit-learn.

Example code:

from tslearn.preprocessing import TimeSeriesScalerMinMax

scaler = TimeSeriesScalerMinMax()
ts_data_scaled = scaler.fit_transform(ts_data)

Example warning:

FutureWarning: 'force_all_finite' was renamed to 'ensure_all_finite' in 1.6 and will be removed in 1.8.

The latest version of scikit-learn at the time of the creation of this issue is 1.7.0.

[charavelg]

Hi, thanks for the feedback

Support of scikit-learn 1.7 is planned but next release will probably support scikit-learn 1.6 only, with hopefully FutureWarning not poping up anymore.

[charavelg]

Support of scikit learn 1.7 is tracked in #549