Does this issue occur when all extensions are disabled?: Not sure (need extensions to use Jupyter)

• VS Code Version: 1.78.2
• OS Version: Windows 11 Enterprise 22H2
• Using Jupyter Notebooks in VS Code
• Launching VS Code from WSL2 Ubuntu 18.04 terminal
• Connecting Jupyter Notebook with conda environment (although issue seems to repro without conda)
• Jupyter extension v2023.4.10112418018

Steps to Reproduce:

1. Open VS code from WSL2 with code . (where conda environments are already initialized)
2. Create a new Jupyter Notebook in VS Code
3. Connect the Jupyter Notebook to a conda environment as the python kernel (select Kernel >> Python Environments >> conda environment)
4. Populate the first python cell of the jupyter notebook with the following code:

   # Prompt the user 10 times (0-9 times) and ask them to enter the answer number
   # this will then print the question number and answer number together
   for question_num in range(0, 10):
       answer_num = input("Enter the Answer Number:")
       print(f"Question Num = #{question_num}; Answer Num = #{answer_num}")

5. Execute the cell and enter your answers as 0, 1, 2, 3, 4, 5, 6, 7, 8, 9 (one at a time). Note that the answer numbers always remain in sync with the question numbers, but the bug shows up when the question-answer pairs aren't always being displayed to the console after providing the answer number. For example:

Here I'm about to enter 3 as an answer after having entered "0", "1", and "2"; Only "0" and "1" have flushed:

Here I just entered 3 and am about to enter 4 and all 4 previous response ("0", "1", "2", "3") have flushed:

Note that this issue doesn't occur on IDEs such as JupyterLab and Jupyter Notebooks, only on VS Code Jupyter Notebooks.

Solution/Workaround:
The solution/workaround was to add a sys.stdout.flush() before prompting the next time around, as shown below:

It would be good however for VS Code to not do this buffering, especially since JupyterLab and Jupyter Notebook don't run into this issue.