Treat .md destinations from PDF as vanilla Markdown

[joliss]

This is the code I'm using to generate vanilla .md files with this tool (issue #14). I don't know whether this is the best way to implement it -- maybe there should be a command line option instead. I'd find this behavior reasonably intuitive though.

The reason I needed vanilla markdown files is because I'm putting the results into LLM prompts, so I want the Markdown to be easy to read and token-efficient.

I'm noticing that further down in mpx.js, converting from .mmd to .md just results in an fs.copy. I'm not sure if there are conversion facilities between .mmd and .md -- if there are, that code might need adjusting too to for consistency.

I also looked at adding vanilla markdown to the image conversion functions, but the outputType ("mmd" here) in calls to ConvertImage isn't used in the implementation:

async function ImageToMMD(source, destination) {
  await ConvertImage("mmd", source, destination);
}

I probably won't have the bandwidth to update this pull request, so if it's not useful, please feel free to just close, or use it as a starting point.