Add Cross-Region Inference for Bedrock Chat Models

[srdas]

Fixes Issue #1114

Fixes PR #1014
(supersedes this PR, which is being closed)

Cross region inference is now supported in Bedrock. In particular this is available for the latest models such as Llama3.2 and Claude-Sonnet-3.5, etc.

This PR adds all Llama3.2 models to the list of available models.

Also added is a cross region inference area field required if a model is being used from another region, which is possible now for models not available in the user's region.

If a model is not available in a region, then using it will generate an error:

File "/opt/anaconda3/envs/jupyter-ai/lib/python3.12/site-packages/langchain_aws/llms/bedrock.py", line 857, in _prepare_input_and_invoke_stream
    raise ValueError(f"Error raised by bedrock service: {e}")
ValueError: Error raised by bedrock service: An error occurred (ValidationException) when calling the InvokeModelWithResponseStream operation: Invocation of model ID meta.llama3-2-3b-instruct-v1:0 with on-demand throughput isn’t supported. Retry your request with the ID or ARN of an inference profile that contains this model.

To avail of cross region inference, enter the area from this page into the field provided:

The model will now be available and will not error out: