refactor: add suggestions from PR

Add suggestsions made by CodeRabbit on pull request

cognee/infrastructure/llm/get_llm_client.py

@@ -22,13 +22,12 @@ def get_llm_client():
 
     # Check if max_token value is defined in liteLLM for given model
     # if not use value from cognee configuration
-    from cognee.infrastructure.llm.utils import get_model_max_tokens
+    from cognee.infrastructure.llm.utils import (
+        get_model_max_tokens,
+    )  # imported here to avoid circular imports
 
-    max_tokens = (
-        get_model_max_tokens(llm_config.llm_model)
-        if get_model_max_tokens(llm_config.llm_model)
-        else llm_config.llm_max_tokens
-    )
+    model_max_tokens = get_model_max_tokens(llm_config.llm_model)
+    max_tokens = model_max_tokens if model_max_tokens else llm_config.llm_max_tokens
 
     if provider == LLMProvider.OPENAI:
         if llm_config.llm_api_key is None:

cognee/modules/data/processing/document_types/Document.py

@@ -11,5 +11,5 @@ class Document(DataPoint):
     mime_type: str
     _metadata: dict = {"index_fields": ["name"], "type": "Document"}
 
-    def read(self, chunk_size: int, chunker=str) -> str:
+    def read(self, chunk_size: int, chunker=str, max_chunk_tokens: Optional[int] = None) -> str:
         pass

cognee/tasks/repo_processor/get_source_code_chunks.py

@@ -97,8 +97,6 @@ def _get_chunk_source_code(
     current_source_code = ""
 
     # Get embedding engine used in vector database
-    from cognee.infrastructure.databases.vector.get_vector_engine import get_vector_engine
-
     embedding_engine = get_vector_engine().embedding_engine
 
     for i, (child_code, token_count) in enumerate(code_token_counts):