Update docstrings to match the current guideline.

Motivation

Recently, we have introduced the docstrings guideline, and it is written in wiki.

However, some docstrings do not follow it. So, here, I would like to make a list which we need to update.

List

Need to add examples

Since optuna users will use these functions and methods, they should have an example to explain how to use it. So, we need to add an Example: section in the docstring.

• optuna.logging.get_verbosity #3066
• optuna.logging.set_verbosity #3061
• optuna.logging.enable_default_handler
• optuna.logging.disable_propagation
• optuna.study.create_study #1711
• optuna.study.load_study #1712
• optuna.study.delete_study #1741
• optuna.study.get_all_study_summaries #1742
• optuna.study.Study.optimize #1726
• optuna.study.Study.get_user_attr
• optuna.study.Study.set_user_attr
• optuna.study.Study.stop #1752
• optuna.study.Study.get_trials #1746
• optuna.multi_objective.study.MultiObjectiveStudy.optimize #1747
• optuna.multi_objective.study.create_study #1749
• optuna.multi_objective.study.load_study #1750
• …

Need to add enough description

Since optuna users will use these functions and methods, they should explain their role and usage. So, we need to add more explanation in the docstring.

• ~optuna.study.Study.get_user_attr~
• optuna.study.Study.user_attrs #1745
• optuna.study.Study.set_user_attr #1744
• …

Need to add see also

Since the following functionalities are related each other, we need to add a .. seealso:: section in the docstring.

• optuna.study.Study.get_trials <-> optuna.study.Study.trials
• …

[JoshKoh]

I think we need examples for multiple objective optimization and study.stop as introduced recently in ver. 1.5. Especially how to stop study once objective function achieves certain threshold. Also, example for pruning needs more explanation: for step in range(100): # what is this for? Why do we need range(100), how to determine a range? clf.partial_fit(train_x, train_y, classes=classes)

    # Report intermediate objective value.
    intermediate_value = 1.0 - clf.score(valid_x, valid_y)
    trial.report(intermediate_value, step)

[hvy]

Style varies quite a bit between example doctest code, e.g. some use single while other use double quotes for strings. Maybe we should make them consistent, following black?