Add Gemma 3n to KerasHub

[harshaljanjani]

It's just the beginning; still a long way to go.
I'm traveling and will be AFK for a while, so I've pushed the work done so far. MobileNetV5's design integrates well with Gemma3n, and we have a complete forward pass on the backbone. There's still a lot to do. For the reviews, if you have any questions or feedback related to the design of MobileNetV5 or Gemma3n so far, feel free to let me know!

Closes #2320.

Colab Notebook

Tracking Tests Until MobileNetV5 Reviews

Checklist

• I have added all the necessary unit tests for my change.
• I have verified that my change does not break existing code and works with all backends (TensorFlow, JAX, and PyTorch).
• My PR is based on the latest changes of the main branch (if unsure, rebase the code).
• I have followed the Keras Hub Model contribution guidelines in making these changes.
• I have followed the Keras Hub API design guidelines in making these changes.
• I have signed the Contributor License Agreement.

[harshaljanjani]

Status on the PR:
DONE: Backbone forward pass with the MobileNetV5 sub-model is working, and all tests (quantization, mixed precision, etc.) are passing.
IN PROGRESS: Task model and conversion script.