Fix global naive models cannot be used in ensemble models

[dennisbader]

Checklist before merging this PR:

• Mentioned all issues that this PR fixes or addresses.
• Summarized the updates of this PR under Summary.
• Added an entry under Unreleased in the Changelog.

Fixes an issue mentioned on gitter where Global Naive models could not be used in ensemble models.

Code to reproduce below.

Summary

• Fixed a bug where global naive models could not be used in ensemble models
• Fixed a bug with global naive models where some supports_* properties were wrongly defined as methods.

Additional Info

Code to reproduce the issue:

import numpy as np

from darts.datasets import AirPassengersDataset
from darts.models import NaiveEnsembleModel, GlobalNaiveDrift

series = AirPassengersDataset().load().astype(np.float32)

model = NaiveEnsembleModel(
    forecasting_models=[
        GlobalNaiveDrift(input_chunk_length=12, output_chunk_length=12).fit(series)
    ],
    train_forecasting_models=False,
)
pred_kwargs = {"predict_likelihood_parameters": True, "verbose": True}
model.predict(n=12, series=series, **pred_kwargs)

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 94.19%. Comparing base (d6371b5) to head (b1a86f4).
Report is 1 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2666      +/-   ##
==========================================
- Coverage   94.25%   94.19%   -0.07%     
==========================================
  Files         141      141              
  Lines       15550    15552       +2     
==========================================
- Hits        14657    14649       -8     
- Misses        893      903      +10     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.