Cleanup cuML ImportError message

[beckernick]

Related Issue or bug

Closes #3422

Describe the changes you've made

This PR updates the ImportError message when using use_gpu="force" and cuML is not available. The message "Install or update cuML to be >= 22.10" is clear enough without trying to reference the current version installed.

It also adds a note about installation instructions.

Type of change

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Code style update (formatting, local variables)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

How Has This Been Tested?

Tested locally by creating a fresh environment and installing PyCaret via pip, via Github (master branch), and via this PR's local branch.

mamba create -n pycaret -c conda-forge python=3.10 jupyterlab ipython
conda activate pycaret
# cd to local branch
python -m pip install .

from pycaret.datasets import get_data
data = get_data('diabetes')

from pycaret.classification import *
s = setup(data, target = 'Class variable', use_gpu="force")
---------------------------------------------------------------------------
ImportError                               Traceback (most recent call last)
Input In [3], in <cell line: 5>()
      2 data = get_data('diabetes')
      4 from pycaret.classification import *
----> 5 s = setup(data, target = 'Class variable', use_gpu="force")

File ~/NVIDIA/pycaret/pycaret/classification/functional.py:603, in setup(data, data_func, target, index, train_size, test_data, ordinal_features, numeric_features, categorical_features, date_features, text_features, ignore_features, keep_features, preprocess, create_date_columns, imputation_type, numeric_imputation, categorical_imputation, iterative_imputation_iters, numeric_iterative_imputer, categorical_iterative_imputer, text_features_method, max_encoding_ohe, encoding_method, rare_to_value, rare_value, polynomial_features, polynomial_degree, low_variance_threshold, group_features, group_names, drop_groups, remove_multicollinearity, multicollinearity_threshold, bin_numeric_features, remove_outliers, outliers_method, outliers_threshold, fix_imbalance, fix_imbalance_method, transformation, transformation_method, normalize, normalize_method, pca, pca_method, pca_components, feature_selection, feature_selection_method, feature_selection_estimator, n_features_to_select, custom_pipeline, custom_pipeline_position, data_split_shuffle, data_split_stratify, fold_strategy, fold, fold_shuffle, fold_groups, n_jobs, use_gpu, html, session_id, system_log, log_experiment, experiment_name, experiment_custom_tags, log_plots, log_profile, log_data, verbose, memory, profile, profile_kwargs)
...
File ~/NVIDIA/pycaret/pycaret/internal/pycaret_experiment/tabular_experiment.py:363, in _TabularExperiment._initialize_setup(self, n_jobs, use_gpu, html, session_id, system_log, log_experiment, experiment_name, memory, verbose)
    360 message = """cuML is outdated or not found. Required version is >=22.10.
    361 Please visit https://rapids.ai/ for installation instructions."""
    362 if use_gpu == "force":
--> 363     raise ImportError(message)
    364 else:
    365     self.logger.warning(message)

ImportError: cuML is outdated or not found. Required version is >=22.10.
                Please visit https://rapids.ai/ for installation instructions.

Please describe the tests that you ran to verify your changes. Provide instructions so we can reproduce.

Describe if there is any unusual behaviour of your code(Write NA if there isn't)

NA

Checklist:

• My code follows the style guidelines of this project.
• I have performed a self-review of my own code.
• I have commented my code, particularly in hard-to-understand areas.
• I have made corresponding changes to the documentation.
• My changes generate no new warnings.
• I have added tests that prove my fix is effective or that my feature works.
• New and existing unit tests pass locally with my changes.
• Any dependent changes have been merged and published in downstream modules.

(Assuming additional tests are not needed).

Screenshots

Original	Updated
original screenshot	updated screenshot

[Yard1]

Thanks!

[celestinoxp]

Windows can´t use cuml package, so we need to do someting about this: Windows should not have this message!
I created a PR to fix warnings about cuml package on windows.
PR: #3881