Deal with markdown outputs of cells

[choldgraf]

We don't currently deal with the case that cells output markdown content, we should do so!

Would we want the markdown to become part of the top-level doctree, or would we want it to remain inside the CellOutputNode container it is a part of?

originally mentioned in #30

[akhmerov]

I think taking these out of the CellOutputNode would make the doctree too confusing.

If the appearance of the cells and outputs is easy to control (e.g. hide any extra boxes or output markers), I hope there wouldn't be a problem.

[choldgraf]

@akhmerov well I think the problem would be things like we had originally ran into in this repository. If the markdown parser finds a heading, then in the page doctree it will create a new section at the top level and start appending to that. So if someone output markdown that had headings in it, then the parser would normally start pulling that out of the CellOutputNode it's supposed to be in.

Maybe if we made the assumption: any document structure that is generated from a code cell should not be a part of the top-level doctree, then we could render that markdown into docutils nodes at transform time using a new Document object, and then just do cell_output_node.children = document.children

[chrisjsewell]

closed with https://myst-nb.readthedocs.io/en/latest/use/formatting_outputs.html#markdown