Native Docling Integration for PDF-to-Markdown/Text conversion (Consumer & Full-Text Indexing)

Description Description: Problem / Motivation Currently, complex PDFs (such as invoices, multi-page documents, or documents containing intricate tables) are processed using standard text extraction tools. While this works for basic full-text search, the resulting plain text often lacks structure, making it suboptimal for downstream RAG (Retrieval-Augmented Generation) applications and LLM context windows. Tools like Docling (by IBM) excel at converting PDFs into structured Markdown (.md) or clean text formats while preserving table structures, reading order, and hierarchy. Proposed Solution I would like to propose a feature that allows Paperless-ngx to leverage external or integrated advanced parsers like Docling for document ingestion: New Consumer Folder / Processing Option: In addition to the standard consume folder, introduce an optional workflow or companion folder (e.g., consume-md or a pre-processing hook) where users can supply pre-converted Markdown files alongs…