fix: add null guard for markdown_processor in read_entity_content

Pyright flagged `self.markdown_processor.read_file()` as
reportOptionalMemberAccess after markdown_processor was made Optional
in the FileService refactor. Added a fail-fast ValueError if
markdown_processor is None when read_entity_content is called —
consistent with the project's fail-fast principle and allows Pyright
to narrow the type for the subsequent call.

Co-authored-by: jope-bm <jope-bm@users.noreply.github.com>
Signed-off-by: claude[bot] <41898282+claude[bot]@users.noreply.github.com>

Author: github-actions[bot]

src/basic_memory/services/file_service.py

@@ -79,6 +79,10 @@ async def read_entity_content(self, entity: EntityModel) -> str:
         """
         logger.debug(f"Reading entity content, entity_id={entity.id}, permalink={entity.permalink}")
 
+        # markdown_processor is required for entity content reads — fail fast if not configured
+        if self.markdown_processor is None:
+            raise ValueError("markdown_processor is required for read_entity_content")
+
         file_path = self.get_entity_path(entity)
         markdown = await self.markdown_processor.read_file(file_path)
         return markdown.content or ""