Export and re-import Neo4j graph state instead of regenerating from scratch for each pipeline run.
- On complete: export graph via CALL apoc.export.cypher.all() to .cypher dump (save in Drive or Blob)
- On startup: import dump if present (skip LLM extraction)
- Add just dump-graph/restore-graph commands
- Track pipeline_run node (timestamp, source file hashes; incremental runs only process new/changed files)
Export and re-import Neo4j graph state instead of regenerating from scratch for each pipeline run.