There a slight mismatch between the input for anchor image (no batch) and integrated gradients (a batch)

[seldondev]

This might be a ticket on the alibi / core side but it is referenced here as it might affect mlserver batching.

Currently we can only do one explanation for some explainers (e.g. AnchorImage), and the call to explain will not allow a batch dimension.

This is different from other explainers (e.g. IntegratedGradients) where there is a batch dimension.

How are we going to deal with this? extra metadata to be passed for the model specifying batch or not batch?