Fix incorrect output from prints originating from different processes

In the PipeFunc documentation I got:
```
```
  {
   "cell_type": "code",
   "execution_count": 47,
   "id": "92",
   "metadata": {
    "execution": {
     "iopub.execute_input": "2024-05-31T05:42:39.297713Z",
     "iopub.status.busy": "2024-05-31T05:42:39.297474Z",
     "iopub.status.idle": "2024-05-31T05:42:40.477462Z",
     "shell.execute_reply": "2024-05-31T05:42:40.475729Z"
    }
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-30 22:42:39.410279 - Running double_it for x=3"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-30 22:42:39.408318 - Running double_it for x=0"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-30 22:42:39.410888 - Running double_it for x=1"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-30 22:42:39.416024 - Running double_it for x=2"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-30 22:42:39.431485 - Running half_it for x=0"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-30 22:42:39.434285 - Running half_it for x=1"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-30 22:42:39.433559 - Running half_it for x=2"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-30 22:42:39.439223 - Running half_it for x=3"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "2024-05-30 22:42:40.459668 - Running take_sum"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "14\n"
     ]
    }
   ],
   "source": [
    "from concurrent.futures import ProcessPoolExecutor\n",
    "import datetime\n",
    "import numpy as np\n",
    "import time\n",
    "from pipefunc import Pipeline, pipefunc\n",
    "\n",
    "\n",
    "@pipefunc(output_name=\"double\", mapspec=\"x[i] -> double[i]\")\n",
    "def double_it(x: int) -> int:\n",
    "    print(f\"{datetime.datetime.now()} - Running double_it for x={x}\")\n",
    "    time.sleep(1)\n",
    "    return 2 * x\n",
    "\n",
    "\n",
    "@pipefunc(output_name=\"half\", mapspec=\"x[i] -> half[i]\")\n",
    "def half_it(x: int) -> int:\n",
    "    print(f\"{datetime.datetime.now()} - Running half_it for x={x}\")\n",
    "    time.sleep(1)\n",
    "    return x // 2\n",
    "\n",
    "\n",
    "@pipefunc(output_name=\"sum\")\n",
    "def take_sum(half: np.ndarray, double: np.ndarray) -> int:\n",
    "    print(f\"{datetime.datetime.now()} - Running take_sum\")\n",
    "    return sum(half + double)\n",
    "\n",
    "\n",
    "pipeline = Pipeline([double_it, half_it, take_sum])\n",
    "inputs = {\"x\": [0, 1, 2, 3]}\n",
    "run_folder = \"my_run_folder\"\n",
    "executor = ProcessPoolExecutor(max_workers=8)  # use 8 processes\n",
    "results = pipeline.map(\n",
    "    inputs,\n",
    "    run_folder=run_folder,\n",
    "    parallel=True,\n",
    "    executor=executor,\n",
    "    storage=\"shared_memory_dict\",\n",
    ")\n",
    "print(results[\"sum\"].output)"
   ]
  },
```

myst_nb/core/utils.py

@@ -25,9 +25,12 @@ def coalesce_streams(outputs: list[NotebookNode]) -> list[NotebookNode]:
     for output in outputs:
         if output["output_type"] == "stream":
             if output["name"] in streams:
-                streams[output["name"]]["text"] += output["text"]
+                out = output["text"].strip()
+                if out:
+                    streams[output["name"]]["text"] += f"{out}\n"
             else:
                 new_outputs.append(output)
+                output["text"] = output["text"].strip() + "\n"
                 streams[output["name"]] = output
         else:
             new_outputs.append(output)