AWS switched to requiring region-prefixed inference profile IDs instead of
foundation model IDs. Added auto-detection based on boto3 client region
to convert model IDs like meta.llama3-1-70b-instruct-v1:0 to
us.meta.llama3-1-70b-instruct-v1:0 depending on the region.
Handles edge cases like ARNs, case sensitivity, and None regions.
This fixes the ValidationException error when using on-demand throughput.
