[Feature] Add Amazon Bedrock integration for chat models and embedding models

[avichaym]

Search before asking

• I searched in the issues and found nothing similar.

Description

Motivation

Flink Agents currently supports Ollama, OpenAI, Anthropic (direct), and Azure AI as chat model providers, and Ollama/OpenAI for embeddings. There is no integration for Amazon Bedrock, which is the primary LLM gateway for AWS customers.

Proposed Changes

Add two new integration modules:

• Chat model (integrations/chat-models/bedrock/) — Uses the Bedrock Converse API with native tool calling support. SigV4 auth via DefaultCredentialsProvider. Supports all Bedrock models accessible via Converse API (Claude, Llama, Mistral, Titan, etc.)

• Embedding model (integrations/embedding-models/bedrock/) — Uses Titan Text Embeddings V2 via InvokeModel. Batch embed(List<String>) parallelizes via configurable thread pool (embed_concurrency parameter, default 4).

I have a working implementation with unit tests ready to submit as a PR.

Are you willing to submit a PR?

• I'm willing to submit a PR!

[xintongsong]

Thanks for offering the contribution, @avichaym. The proposal sounds good to me. I've assigned you to the ticket. Please move ahead.