refactor(api): move llm quota deduction to app graph layer (#32786)

api/core/rag/index_processor/processor/paragraph_index_processor.py

@@ -8,6 +8,7 @@ from typing import Any, cast
 
 logger = logging.getLogger(__name__)
 
+from core.app.llm import deduct_llm_quota
 from core.entities.knowledge_entities import PreviewDetail
 from core.llm_generator.prompts import DEFAULT_GENERATOR_SUMMARY_PROMPT
 from core.model_manager import ModelInstance
@@ -35,7 +36,6 @@ from core.rag.models.document import AttachmentDocument, Document, MultimodalGen
 from core.rag.retrieval.retrieval_methods import RetrievalMethod
 from core.tools.utils.text_processing_utils import remove_leading_symbols
 from core.workflow.file import File, FileTransferMethod, FileType, file_manager
-from core.workflow.nodes.llm import llm_utils
 from extensions.ext_database import db
 from factories.file_factory import build_from_mapping
 from libs import helper
@@ -474,7 +474,7 @@ class ParagraphIndexProcessor(BaseIndexProcessor):
 
         # Deduct quota for summary generation (same as workflow nodes)
         try:
-            llm_utils.deduct_llm_quota(tenant_id=tenant_id, model_instance=model_instance, usage=usage)
+            deduct_llm_quota(tenant_id=tenant_id, model_instance=model_instance, usage=usage)
         except Exception as e:
             # Log but don't fail summary generation if quota deduction fails
             logger.warning("Failed to deduct quota for summary generation: %s", str(e))