Don't run ipynb-filters for qmd source files

The idea behind ipynb-filters is that you have a notebook created for some other purpose / within some other system that you want to adapt for processing by quarto.

However, if you are creating a qmd no such adaptation is required because the file is natively targeting quarto (and can be filtered as required with standard pandoc filters)

news/changelog-1.1.md

@@ -4,6 +4,7 @@
 - Refine over-detection of Jupyter shell magics (which prevented kernel daemonization)
 - Use on-disk cache for filtered jupyter notebooks
 - Prevent error in `quarto check` when Jupyter is installed but has no Python kernel
+- Don't run `ipynb-filters` for qmd source files (only run them for ipynb source files)
 
 ## Knitr
 

src/execute/jupyter/jupyter.ts

@@ -271,10 +271,13 @@ export const jupyterEngine: ExecutionEngine = {
       }
     }
 
-    // convert to markdown and write to target
+    // convert to markdown and write to target (only run notebook filters
+    // if the source is an ipynb file)
     const nbContents = await jupyterNotebookFiltered(
       options.target.input,
-      options.format.execute[kIpynbFilters],
+      isJupyterNotebook(options.target.source)
+        ? options.format.execute[kIpynbFilters]
+        : [],
     );
     const nb = jupyterFromJSON(nbContents);
     const assets = jupyterAssets(