Implement multitask support in multimodal API

Current API does not support multitasking on multimodal data. However, the probable user may want to solve two tasks (regression and classification) simultaneously.

So this opportunity should be added like in this [example](https://github.com/nccr-itmo/FEDOT/blob/master/examples/simple/multitask_classification_regression_api.py).
The corresponding changes in core logic, validation rules, and test should be implemented.