feat: Add a plot to ComparisonReport summarize display

[MarieSacksick]

The plot of this display will be a pair plot to visualize one metric against another.

This is part of the narrative "I have several models, how can I choose the best one?".

Closes #1340

[github-actions]

Coverage Report for skore/

File	Stmts	Miss	Cover	Missing
skore/src/skore
__init__.py	23	0	100%
_config.py	31	0	100%
exceptions.py	4	4	0%	4, 15, 19, 23
skore/src/skore/_sklearn
__init__.py	6	0	100%
_base.py	198	14	92%	45, 58, 127, 130, 183, 186–187, 189–192, 225, 228–229
find_ml_task.py	61	0	100%
types.py	27	1	96%	28
skore/src/skore/_sklearn/_comparison
__init__.py	7	0	100%
feature_importance_accessor.py	39	2	94%	90, 109
metrics_accessor.py	183	4	97%	174, 188, 271, 1233
report.py	107	0	100%
utils.py	54	0	100%
skore/src/skore/_sklearn/_cross_validation
__init__.py	9	0	100%
data_accessor.py	45	3	93%	134, 137, 140
feature_importance_accessor.py	24	0	100%
metrics_accessor.py	182	1	99%	249
report.py	135	1	99%	487
skore/src/skore/_sklearn/_estimator
__init__.py	9	0	100%
data_accessor.py	66	1	98%	82
feature_importance_accessor.py	144	2	98%	223–224
metrics_accessor.py	356	8	97%	200, 202, 209, 300, 369, 373, 388, 423
report.py	167	2	98%	448–449
skore/src/skore/_sklearn/_plot
__init__.py	3	0	100%
base.py	102	7	93%	61–62, 200, 224–226, 230
utils.py	80	0	100%
skore/src/skore/_sklearn/_plot/data
__init__.py	2	0	100%
table_report.py	185	1	99%	706
skore/src/skore/_sklearn/_plot/metrics
__init__.py	6	0	100%
confusion_matrix.py	70	4	94%	92, 100, 122, 230
feature_importance_display.py	67	21	68%	88, 121–122, 124, 142–146, 148–155, 158–160, 162
metrics_summary_display.py	73	0	100%
precision_recall_curve.py	280	5	98%	455, 555, 559, 619, 751
prediction_error.py	227	5	97%	179, 186, 422, 505, 705
roc_curve.py	292	8	97%	385, 508, 513, 614, 619, 623, 692, 832
skore/src/skore/_sklearn/train_test_split
__init__.py	0	0	100%
train_test_split.py	58	0	100%
skore/src/skore/_sklearn/train_test_split/warning
__init__.py	8	0	100%
high_class_imbalance_too_few_examples_warning.py	19	1	94%	83
high_class_imbalance_warning.py	20	0	100%
random_state_unset_warning.py	10	0	100%
shuffle_true_warning.py	9	0	100%
stratify_is_set_warning.py	10	0	100%
time_based_column_warning.py	21	0	100%
train_test_split_warning.py	3	0	100%
skore/src/skore/_utils
__init__.py	6	2	66%	8, 13
_accessor.py	90	3	96%	34, 146, 190
_environment.py	27	1	96%	40
_fixes.py	8	0	100%
_index.py	5	0	100%
_logger.py	22	4	81%	15–17, 19
_measure_time.py	10	0	100%
_parallel.py	38	3	92%	23, 33, 124
_patch.py	13	5	61%	21, 23–24, 35, 37
_progress_bar.py	46	0	100%
_repr_html.py	8	0	100%
_show_versions.py	38	0	100%
_testing.py	55	0	100%
skore/src/skore/project
__init__.py	2	0	100%
project.py	48	0	100%
summary.py	75	1	98%	120
widget.py	189	0	100%
TOTAL	4102	114	97%

Tests	Skipped	Failures	Errors	Time
1086	5 💤	0 ❌	0 🔥	4m 22s ⏱️

[github-actions]

Documentation preview @ 98f4829

[glemaitre]

It is a quick review. A couple of additional thoughts:

• we will want to have some logspace sometimes. In scikit-learn, we came with an heuristic with the time for the validation curve: https://github.com/scikit-learn/scikit-learn/blob/d4d4af8c4/sklearn/model_selection/_plot.py#L111-L117 I think that we can retake this implementation
• we need to put some thoughts regarding score vs. loss.
• I think that we need to tackle the following issue first: #1837

[MarieSacksick]

we will want to have some logspace sometimes. In scikit-learn, we came with an heuristic with the time for the validation curve: https://github.com/scikit-learn/scikit-learn/blob/d4d4af8c4/sklearn/model_selection/_plot.py#L111-L117 I think that we can retake this implementation

About this, it looks like an improvement to add this option imo. It could be a good first issue. Since we have a workshop soon, what keeping it for this workshop?

I think that we need to tackle the following issue first: #1837

If you think that this other issue can be done soon, let's wait indeed! Otherwise, I don't really like keeping PR open for a while, we tend to forget the code and the context.

we need to put some thoughts regarding score vs. loss.

Do you have the pareto front in mind?
As the first feedback, to me it's an improvement, not light this one, and I prefer to do step by step and releasing this first part when we are ready.

[glemaitre]

About this, it looks like an improvement to add this option imo. It could be a good first issue. Since we have a workshop soon, what keeping it for this workshop?

It is 5 lines of code already tested somewhere else. I would not bother since I already know that we want it.

Do you have the pareto front in mind?
As the first feedback, to me it's an improvement, not light this one, and I prefer to do step by step and releasing this first part when we are ready.

Not really, it was more on the direction of the loss/score. But I think we can just align the implementation with the display on the hub.

In terms of implementation, I don't like a partial implementation.

[MarieSacksick]

@glemaitre, can you be more precise about your expectations about the following sentence please?

Not really, it was more on the direction of the loss/score. But I think we can just align the implementation with the display on the hub.

So far, the only difference with the hub is the display of the pareto front and the colored aspect on the "worse" side of the front.

[MarieSacksick]

Ready for a new round of review @glemaitre :) ! (still in draft because still waiting for your PR first).