Enhance MeanDice metric

[yiheng-wang-nv]

In the validation step of the hovernet pipeline (https://github.com/vqdang/hover_net/blob/842964dc5d26ffe196d236d16fad16643a103850/models/hovernet/run_desc.py#L282-L290), dice is calculated by accumulating intersections and total sections of all predictions and ground truth pairs, and then do the
division.

cc @KumoLiu @drbeh @dongyang0122 @daguangxu @wyli @Nic-Ma

[bhashemian]

Hi @yiheng-wang-nv, could you please let us know what is the status of this ticket? Are you working on a PR to address it?

[yiheng-wang-nv]

Hi @yiheng-wang-nv, could you please let us know what is the status of this ticket? Are you working on a PR to address it?

Hi @drbeh , I will start doing this feature.

[yiheng-wang-nv]

Hi @yiheng-wang-nv, could you please let us know what is the status of this ticket? Are you working on a PR to address it?

Hi @drbeh , I will start doing this feature.

Hi @drbeh , btw, do you have any suggestions on the name of this metric?
In the previously closed PR, I finally decided to implement a sub class rather than modify the class DiceMetric. Due to other more urgent tasks, I did not do the following changes till now.