Fix stream output

[davidbrochart]

Fixes #293.

Currently, the frontend expects a stream output "text" field to be a YText:

jupyter_ydoc/javascript/src/ycell.ts

Lines 791 to 799 in b208786

	if (output.output_type === 'stream') {
	// Set the text field as a Y.Text
	const { text, ...outputWithoutText } = output;
	_newOutput1 = outputWithoutText;
	const newText = new Y.Text();
	let _text = text instanceof Array ? text.join() : (text as string);
	newText.insert(0, _text);
	_newOutput1['text'] = newText;
	} else {

But the backend expects it to be a YArray:

jupyter_ydoc/jupyter_ydoc/ynotebook.py

Lines 165 to 166 in b208786

	if output.get("output_type") == "stream":
	output["text"] = Array(output.get("text", []))

Creating a notebook with:

text = "*" * 10_000

nb = """
{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "fa183e4b-bdfc-4789-a2ff-6ff197e31f3c",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "TEXT"
     ]
    }
   ],
   "source": [
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.13.0"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}
""".replace("TEXT", text)

with open("Untitled.ipynb", "w") as f:
    f.write(nb)

And opening it in JupyterLab takes a few seconds and shows the following output:

*
*
*
...

Which is wrong because it should be one single line.
With this PR, the notebook opens instantly, showing a single line:

***...

@trungleduc I remember you updated the snapshots somewhere because you saw a mismatch like that. Was it in jupyter-collaboration?

[trungleduc]

Yes, it's here jupyterlab/jupyter-collaboration#396 (comment). This PR should also close jupyterlab/jupyterlab#16934 ?

[davidbrochart]

Yes, thanks for the links.