feat(config): move embedding config to llmConfiguration.embeddings; slim naturalLanguageSearch

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

Author: pmbrull

conf/openmetadata.yaml

@@ -590,33 +590,8 @@ elasticsearch:
   naturalLanguageSearch:
     enabled: ${NATURAL_LANGUAGE_SEARCH_ENABLED:-false}
     semanticSearchEnabled: ${SEMANTIC_SEARCH_ENABLED:-false}
-    embeddingProvider: ${EMBEDDING_PROVIDER:-bedrock} # Options: "openai", "bedrock", "google", "djl"
-    maxConcurrentRequests: ${MAX_CONCURRENT_EMBEDDING_REQUESTS:-10}
     providerClass: ${NATURAL_LANGUAGE_SEARCH_PROVIDER_CLASS:-org.openmetadata.service.search.nlq.NoOpNLQService}
-    bedrock:
-      awsConfig:
-        enabled: ${BEDROCK_AWS_IAM_AUTH_ENABLED:-false}
-        region: ${AWS_BEDROCK_REGION:-""}
-        accessKeyId: ${AWS_BEDROCK_ACCESS_KEY:-""}
-        secretAccessKey: ${AWS_BEDROCK_SECRET_KEY:-""}
-        sessionToken: ${AWS_BEDROCK_SESSION_TOKEN:-""}
-      embeddingModelId: ${AWS_BEDROCK_EMBED_MODEL_ID:-""}
-      embeddingDimension: ${AWS_BEDROCK_EMBEDDING_DIMENSION:-""}
-    openai:
-      apiKey: ${OPENAI_API_KEY:-""}
-      # For Azure OpenAI, set endpoint and deploymentName:
-      endpoint: ${OPENAI_API_ENDPOINT:-""}  # e.g., https://your-resource.openai.azure.com
-      deploymentName: ${OPENAI_DEPLOYMENT_NAME:-""}  # Required for Azure OpenAI
-      apiVersion: ${OPENAI_API_VERSION:-"2024-02-01"}  # Azure OpenAI API version
-      embeddingModelId: ${OPENAI_EMBEDDING_MODEL_ID:-"text-embedding-3-small"}
-      embeddingDimension: ${OPENAI_EMBEDDING_DIMENSION:-1536}
-    google:
-      apiKey: ${GOOGLE_API_KEY:-""}  # API key from Google AI Studio
-      embeddingModelId: ${GOOGLE_EMBEDDING_MODEL_ID:-"gemini-embedding-001"}
-      embeddingDimension: ${GOOGLE_EMBEDDING_DIMENSION:-768}  # Sent as outputDimensionality. gemini-embedding-001 supports 768/1536/3072; text-embedding-004 supports 768.
-      endpoint: ${GOOGLE_API_ENDPOINT:-""}  # Optional override; full :embedContent URL. Leave empty to use the default Generative Language API endpoint.
-    djl:
-      embeddingModel: ${DJL_EMBEDDING_MODEL:-"ai.djl.huggingface.pytorch/sentence-transformers/all-MiniLM-L6-v2"}
+    # Embedding provider/model/credentials now live under llmConfiguration.embeddings (single LLM config home).
 
 # Platform-wide LLM provider configuration. Generic - consumed by any feature that needs LLM
 # completions (e.g. Context Center knowledge-pill extraction and the MCP Chat application). The
@@ -651,6 +626,23 @@ llmConfiguration:
     modelId: ${LLM_ANTHROPIC_MODEL_ID:-"claude-3-5-sonnet-20240620"}
     baseUrl: ${LLM_ANTHROPIC_BASE_URL:-"https://api.anthropic.com"}
     maxTokens: ${LLM_ANTHROPIC_MAX_TOKENS:-4096}
+  # Vector embeddings for semantic search. Credentials reuse the provider blocks above
+  # (bedrock.awsConfig, openai.apiKey/endpoint, google.apiKey); this section only selects the
+  # embedding provider + model. The embedding provider may differ from the chat provider above.
+  embeddings:
+    provider: ${EMBEDDING_PROVIDER:-bedrock}  # bedrock | openai | google | djl
+    maxConcurrentRequests: ${MAX_CONCURRENT_EMBEDDING_REQUESTS:-10}
+    bedrock:
+      embeddingModelId: ${AWS_BEDROCK_EMBED_MODEL_ID:-"amazon.titan-embed-text-v2:0"}
+      embeddingDimension: ${AWS_BEDROCK_EMBEDDING_DIMENSION:-512}
+    openai:
+      embeddingModelId: ${OPENAI_EMBEDDING_MODEL_ID:-"text-embedding-3-small"}
+      embeddingDimension: ${OPENAI_EMBEDDING_DIMENSION:-1536}
+    google:
+      embeddingModelId: ${GOOGLE_EMBEDDING_MODEL_ID:-"gemini-embedding-001"}
+      embeddingDimension: ${GOOGLE_EMBEDDING_DIMENSION:-768}
+    djl:
+      embeddingModel: ${DJL_EMBEDDING_MODEL:-"ai.djl.huggingface.pytorch/sentence-transformers/all-MiniLM-L6-v2"}
 
 eventMonitoringConfiguration:
   eventMonitor: ${EVENT_MONITOR:-prometheus}  # Possible values are "prometheus", "cloudwatch"