RMarkdown support

[AaronVerDow]

I was looking into adding language support for RMarkdown but I'm not sure how it should be handled. RMarkdown contents by default are comments and the code chunks (of multiple languages) are quoted. I took a look at Jupyter, which behaves similarly, and found an unusual test format plus a special parse_jupyter function in the source.

What would be the recommended path to add RMarkdown support? Should it be handled as a one-off like Jupyter or could they be combined with their differences defined in languages.json?