[RFC] Remove some deprecated features in v2.0.0.

[toshihikoyanase]

Background

Optuna has deprecated features such as Trial.trial_id and Study.study_id. I think it's a good opportunity to revise them at the major release. Here's a proposal of policy to remove them. If you have any comments or suggestions, please let me know.

Policy to remove deprecated features

• Deprecated features will be removed in the major release after the next in general. For example, the step argument of Trial.should_prune was deprecated in v0.12.0 and will be removed v2.0.0.
• Features that may affect many users can be kept exceptionally. For instance, optuna.structs.TrialPruned was deprecated in v0.19.0 but will be kept in v2.0.0.
• Deprecated features of classes/functions which are marked as experimental will be removed when they promote to non-experimental.

The deprecated features removed in v2.0.0

Feature	Deprecated at	Experimental?
optuna.study.StudySummary.study_id	0.20.0	N
optuna.integration.ChainerMNTrial.trial_id	0.19.0?	N
optuna.integration.LightGBMTuner.__init__(best_params)	1.4.0	Y
optuna.integration.LightGBMTuner.__init__(tuning_history)	1.4.0	Y
optuna.pruners.Hyperband.__init__(n_brackets)	1.4.0	Y
optuna.pruners.Hyperband.__init__(min_early_stopping_rate_low)	1.4.0	Y
optuna.structs.FrozenTrial.trial_id	0.19.0	N
optuna.structs.StudySummary.study_id	0.20.0	N
optuna.study.BaseStudy.storage	0.15.0	N
optuna.study.Study.study_id	0.20.0	N
optuna.trial.FrozenTrial.trial_id	0.19.0?	N
optuna.trial.Trial.sould_prune(step)	0.12.0	N
optuna.trial.Trial.trial_id	0.19.0?	N
optuna.trial.Trial.study_id	0.20.0	N

• 0.20.0 seems to be a typo for 0.19.0 or 1.0.0.
• The deprecation versions of some trial_id attributes were not recorded in the code.
• optuna.integration.LightGBMTuner.best_booster will not be removed in v2.0.0 because some users still use it.
• optuna.structs.TrialPruned will not be removed in v2.0.0 because it had been used in the official examples for a long time.

[toshihikoyanase]

We removed all the deprecated features listed in #1346 (comment) from the master branch. If you have any problems, please let us know.