Test snapshots fail on main - spurious line number is added to first cell

[krassowski]

Noticed in #228

This started failing a week ago in https://github.com/JupyterEverywhere/jupyterlite-extension/actions/runs/17649802454/job/50308498161

[krassowski]

The tests were failing in #85 too (https://github.com/JupyterEverywhere/jupyterlite-extension/actions/runs/17648300767/job/50153514902), probably should not have been merged with failing tests.

[krassowski]

Reverting #85 would indeed solve the regression with code cells including line numbers:

I don't think we should revert, but it at least clarifies what lead to this regression.

[krassowski]

I do not see the line numbers on the GitHub preview nor locally using the production build. I will next test the test build to see if there is a difference.

[krassowski]

I am able to reproduce the issue locally with the test build: