(feat) Integrate Docling for documents conversion

[tobitege]

What problem or use case are you trying to solve?

Integrate IBM Deep Search' "Docling" (MIT) for e.g. agent skills to read/convert multitude of
documents (file/URL) to e.g. markdown:

https://github.com/DS4SD/docling

From their examples:

from docling.document_converter import DocumentConverter

source = "https://arxiv.org/pdf/2408.09869"  # document per local path or URL
converter = DocumentConverter()
result = converter.convert(source)
print(result.document.export_to_markdown())  # output: "## Docling Technical Report[...]"

If a PDF was converted to markdown before sending it to the LLM (maybe as tool calls),
it could bring down cost and/or open up other uses for the results.