Feature/metrics modality support

This PR adds

pixel flipping metrics: PixelFlipping, MoRF, LeRF, and AbPC
a class for post processing in AutoExperiment: PostProcessor
methods to run experiment by batch to Experiment: run_batch, explain_batch, postprocess_batch, evaluate_batch
methods to browse results of experiment to ExperimentManager
- get_data_by_id, get_explainer_by_id, get_postprocessor_by_id, get_metric_by_id
- batch_data_by_ids, batch_explanations_by_ids, batch_evaluations_by_ids
- You can see the example usage of them in tutorials/gradio_auto_expr_imagenet_example.py

OpenXAIProject / pnpxai