[FEAT] [CV] Add support for image detection with segmentation

Is your feature request related to a problem? Please describe. Add support for image detection with segmentation

Describe the solution you'd like We just need a more general IoU and area functions and a new data type that inherits from detection data with a new label and prediction validation

Describe alternatives you've considered A custom data object

Use the existing custom segmentation defined in plot_custom_task_tutorial to:

• [ ] Extend the IoU-based detection metrics to work for segmentation - main issue is defining IoU such that it works for segmentation maps - 4SP
• [ ] Add a VisionData based on MyCustomSegmentationData to the package core, add format validations and everything else needed for segmentation support (format definition, validation, tests) - 2SP
• [ ] Add an example of using deepchecks for segmentation. - 1 SP
• [ ] Add segmentation specific metrics - 2SP

We'll start by doing the first two tasks

1. Start from 2nd bullet - create infra
2. Continue with 4th bullet - Create Dice metric (prediction X true)
3. Plan next steps

Continued by #1888