feat: Add GraphQL service for searching and filtering concepts

[filiperochalopes]

Summary

This PR introduces a new GraphQL API endpoint to the OCL API, enabling efficient searching and filtering of concepts within sources. The implementation leverages Elasticsearch for fast text-based queries while providing a robust fallback to database searches.

Changes

• New Django App: Added core/graphql/ with complete GraphQL schema and resolvers
• GraphQL Schema: Implemented conceptsFromSource query supporting:
• Fetching concepts by specific IDs with optional pagination
• Text-based search with Elasticsearch integration
• Automatic fallback to database search when Elasticsearch is unavailable
• Version resolution for sources (HEAD or specific versions)
• Mapping relationships included in responses
• Dependencies: Added Strawberry GraphQL libraries (strawberry-graphql, strawberry-graphql-django)
• URL Integration: Added GraphQL endpoint at /graphql/ with GraphiQL interface enabled
• Testing: Comprehensive test suite covering pagination, search ordering, error handling, and fallback
scenarios
• Infrastructure: Updated Docker Compose configuration and added documentation

Technical Details

• Uses async resolvers for optimal performance
• Implements custom ordering based on Elasticsearch relevance scores
• Includes proper error handling and GraphQL validation
• Follows existing codebase patterns for source/version resolution and data serialization
• Maintains compatibility with existing REST API endpoints

Verification Steps

1. Start the API with docker compose up
2. Access GraphiQL (Playground) at http://localhost:8000/graphql/
3. Test queries like:

query {
  conceptsFromSource(org: "CIEL", source: "CIEL", query: "hypertension") {
    totalCount
    results {
      conceptId
      display
      mappings { mapType toCode }
    }
  }
}

4. Run tests: docker compose run --rm api python manage.py test core.graphql.tests