Align KP Detection Validation with Model API

[kprokofi]

Summary

This PR removes gap between Model API postprocessing and OTX.
Please note, that to be able to run OV IR inference we will need to support list[Tensor] in TorchDataBatch. This behavior will be removed once we separate OV Models from core OTX Models.

How to test

Checklist

• I have added unit tests to cover my changes.​
• I have added integration tests to cover my changes.​
• I have ran e2e tests and there is no issues.
• I have added the description of my changes into CHANGELOG in my target branch (e.g., CHANGELOG in develop).​
• I have updated the documentation in my target branch accordingly (e.g., documentation in develop).
• I have linked related issues.

License

• I submit my code changes under the same Apache License that covers the project.
  Feel free to contact the maintainers if that's a concern.
• I have updated the license header for each file (see an example below).

# Copyright (C) 2025 Intel Corporation
# SPDX-License-Identifier: Apache-2.0

[kprokofi]

@sovrasov @ashwinvaidya17 please review

[kprokofi]

COCO-person-small

Torch metric:
test/PCK │ 0.723048746585846
OV:
test/PCK │ 0.7259898781776428

Before this change accuracy drop was ~10%