fix!(ipynb writer): improve round trip identity

for raw cell output

BREAKING CHANGE:
The Jupyter ecosystem, including nbconvert, lab and notebook,
deviated from their own spec in nbformat,
where they used the key `raw_mimetype` instead of `format`.

Moreover, the mime-type of rst used in Jupyter
deviated from that suggested by
https://docutils.sourceforge.io/FAQ.html
and is defined as `text/restructuredtext`
when chosen from "Raw NBConvert Format" in Jupyter.

So while this is backward-compatible,
it should matches the real world usage better,
hence improving the round-trip "identity" in raw-cell.

See #229, jupyter/nbformat#229.

src/Text/Pandoc/Writers/Ipynb.hs

@@ -147,7 +147,7 @@ extractCells opts (Div (_id,classes,kvs) xs : bs)
                   "revealjs" -> "text/html"
                   "latex"    -> "text/latex"
                   "markdown" -> "text/markdown"
-                  "rst"      -> "text/x-rst"
+                  "rst"      -> "text/restructuredtext"
                   "asciidoc" -> "text/asciidoc"
                   _          -> f
           (Ipynb.Cell{
@@ -155,7 +155,7 @@ extractCells opts (Div (_id,classes,kvs) xs : bs)
             , cellSource = Source $ breakLines raw
             , cellMetadata = if format' == "ipynb" -- means no format given
                                 then mempty
-                                else M.insert "format"
+                                else M.insert "raw_mimetype"
                                        (Aeson.String format') mempty
             , cellAttachments = Nothing } :) <$> extractCells opts bs
         _ -> extractCells opts bs