[Dallas98]

This pull request introduces a new Helm chart for deploying Label Studio with PostgreSQL on Kubernetes, along with improvements to Milvus configuration and updates to Python dependencies. The most significant changes are the addition of the Label Studio Helm chart (with deployment, service, ingress, persistent storage, and PostgreSQL templates), refactoring Milvus connection configuration for consistency, and updating Python dependencies for the backend.

Label Studio Helm Chart Addition

• Added a new Helm chart for Label Studio, including Chart.yaml, deployment, service, ingress, PostgreSQL deployment/service/pvc, and persistent volume claim templates, enabling Kubernetes-native deployment and configuration. [1] [2] [3] [4] [5] [6] [7]

Milvus Configuration Refactor

• Refactored Milvus connection configuration in MilvusService.java to use a single milvus-uri property, simplifying host/port management.
• Updated Helm deployment and values files to use the new milvus-uri environment variable, ensuring consistent configuration across environments. [1] [2]

Python Backend Dependency Updates

• Updated Python dependencies in pyproject.toml to newer versions and added docx2txt, jq, and adjusted versions for openai, langchain, fastapi, and others for improved compatibility and functionality.
• Added a step to download NLTK data (punkt_tab) in the backend Python Dockerfile for natural language processing support.