table in readme to solve confusion between target and atack model with purpose and data they have seen

[Friedrich-Mueller]

also what training arguments the users/functions have access to

[marisanest]

As far as I understood the task, the requested tables in the README could look like the following. What do you think @Friedrich-Mueller? For the MembershipInferenceBlackBoxRuleBasedAttack and MembershipInferenceAttackOnPointBasis I am not completely sure if this is correct but both attacks do not have an attack_model (correct me if I am wrong) and thus the attack_models do not have seen any data. Of course, we would need to explain the table a bit more.

For the MembershipInferenceBlackBoxAttack:

	training data	test data
target_model	yes	no
attack_model	yes	yes

For the MembershipInferenceBlackBoxRuleBasedAttack:

	training data	test data
target_model	yes	no
attack_model	no	no

For the MembershipInferenceLabelOnlyDecisionBoundaryAttack:

	training data	test data
target_model	yes	no
attack_model	yes	yes

For the MembershipInferenceAttackOnPointBasis:

	training data	test data
target_model	yes	no
attack_model	no	no