Automatically disconnect local conda kernel and switch to dummy kernel.

[fangning-ren]

Environment data

• VS Code version: 1.68.1
• Jupyter Extension version (available under the Extensions sidebar): v2022.5.1001601848
• Python Extension version (available under the Extensions sidebar): v2022.8.0
• OS (Windows | Mac | Linux distro) and version: windows 10 education
• Python and/or Anaconda version: 3.9.12/4.12.0
• Type of virtual environment used: conda
• Jupyter server running: Local

Expected behaviour

connect to my conda kernel and run every cells

Actual behaviour

##Three possible behaviors

1. shows no kernel and do nothing
2. shows a kernel named "Python 3" with no python installed. Click run cell will norify me to install python
3. shows the correct kernel at the beginning. but it switch back to no kernel or the "Python 3" kernel in one seconds.
4. shows the correct kernel at the beginning. Could run a few commands. But soon switch to no kernel or the "Python 3" kernel.

Steps to reproduce:

1. Create a new notebook
2. Open it with vscode.
3. create a new cell.
4. Select kernel to anything except "Python 3".
5. run the cell. The cell may be correctly executed.
6. Then the kernel is autometicly disconnected with no reason.

[DonJayamanne]

Thanks for reporting this issue, i believe this has been resolve din our pre-reelase build.
Please could you:

• Install our pre-release build as follows:
• Go into the extensions tab & search for Jupyter, then click the button Switch to Pre-Release Version
  See below
  

Please let me know how that goes

[amunger]

closing since it has been over a month since the info was requested