ChatBedrock does not support deepseek R1

[efriis]

Bedrock supports deepseek R1 through the invoke API: https://aws.amazon.com/blogs/machine-learning/deepseek-r1-model-now-available-in-amazon-bedrock-marketplace-and-amazon-sagemaker-jumpstart/

A case should be added to the Invoke compatible ChatBedrock case that uses the deepseek message templating (from the above blog):

ChatTemplate.DEEPSEEK: f"""You are a helpful assistant <｜User｜>{prompt}<｜Assistant｜>"""

[efriis]

cc @3coins had someone ask for this today if easy to add!

[efriis]

Would be great to get the Qwen and Llama distill ones working too! There's a chance the llama ones will fall into llama2 chat message templating if the model ID doesn't explicitly include "llama3"

https://github.com/langchain-ai/langchain-aws/blob/main/libs/aws/langchain_aws/chat_models/bedrock.py#L354

[3coins]

Would be great to get the Qwen and Llama distill ones working too

+1, will bring this up with the team. @michaelnchin is working on the deepseek integration in #353.