Hello. I am currently implementing Elasticsearch VectorStore in the Java version of Flink Agents; I intend to create a pull request shortly. Are there any core features that must be implemented? I have primarily referenced the Python version of ChromaVectorStore and also consulted the design documentation extensively. https://github.com/apache/flink-agents/discussions/143

I think for now, it would be sufficient to keep the consistent with the python version on core features, and we can proceed with unified extensions in the future.