Extraction API for Markdown documents

### Question

I have been trying to use the Extraction API and from what I have found it has only been implemented for extraction from PDFs and IMGs. I am trying to use it for a Markdown file by supplying the [_get_default_extraction_option](https://github.com/docling-project/docling/blob/ae61d640c1af7ba584fcbdb775e3f96c7b488531/docling/document_extractor.py#L66) function with the Markdown backend and a custom Markdown pipeline.

I've adapted the VLM extraction pipeline to process markdown files by creating a workaround for the image-based interface. Since it expects a visual input but I need to process plain text markdown, I read the markdown content directly and create a 1x1 pixel dummy image while having all the actual markdown text within the prompt. This lets me use the Markdown file directly and still use most of the pipeline architecture but it feels like a very inefficient solution. Are there plans for implementing the extraction for other file types than just PDFs and IMGs or, if not, is there a better way for me to use the Extraction API for markdown/docs/etc. without such workarounds, since I have found that the accuracy (using the same LLM) is better even with my 'crude' solution?