Implement HuggingFace Object Detection Estimators

[f4str]

Is your feature request related to a problem? Please describe.
Previously, the HuggingFaceClassifierPyTorch was implemented for the classification task. A similar estimator should be implemented for the object detection task. This will allow usage of HuggingFace object detection models like YOLOS, YOLOP, and DETR.

Describe the solution you'd like
A new estimator named HuggingFaceObjectDetector should be implemented under art.estimators.object_detection. The estimator will subclass the PyTorchObjectDetector since it should use the same functionality. It will be structured in a similar way to the HuggingFaceClassifierPyTorch where it takes in the HuggingFace model and preprocessor.

Describe alternatives you've considered
There is consideration whether to call the estimator HuggingFaceObjectDetector or HuggingFaceObjectDetectorPyTorch. Ultimately it is better to use HuggingFaceObjectDetector since most HuggingFace models are PyTorch-based and this name is more concise.

Additional context
This issue will be addressed after #2267 is resolved since it depends on the rebase.