`Add LayoutLMv3 Model`

[carrycooldude]

I would like to contribute the LayoutLMv3 model to KerasHub. This model is a multimodal document understanding model that combines text and image information. I will follow the KerasHub contribution guidelines and provide the necessary code, tests, and Colab notebooks.

[carrycooldude]

This is my PR: #2178