Refactor MaskRCNN rotated detection

[eugene123tw]

Summary

• Align MaskRCNN instance segmentation recipes with MaskRCNN rotated detection recipes.
• Add tile-based recipes for rotated object detection.

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) 2024 Intel Corporation
# SPDX-License-Identifier: Apache-2.0

[kprokofi]

@chuneuny-emily , I think we may merge it before CF and target OTX2.2, since there are no much changes, mostly configs.

[sungchul2]

@eugene123tw @kprokofi Just a question, if rotated detection models use instance segmentation modules, isn't this task a sub task of instance segmentation like multi class classification of classification?

[eugene123tw]

@eugene123tw @kprokofi Just a question, if rotated detection models use instance segmentation modules, isn't this task a sub task of instance segmentation like multi class classification of classification?

Currently, rotated detection can be a subset of inst-seg, as both tasks share the same model interface at the moment (not sure if we are going to add detection based rotated algo), differing in only prediction functions. I guess this relationship makes it somewhat similar to how multi-class classification relates to general classification tasks.