Getting timestamp from Python is flaky

[fcollonval]

Description

I hit the following error with the latest version:

thread '<unnamed>' panicked at src/undo.rs:26:38:
Error getting timestamp: PyErr { type: <class 'SystemError'>, value: SystemError('<built-in function time_ns> returned a result with an exception set'), traceback: Some(<traceback object at 0x7fd05aaf0980>) }

Would it be possible to make that behavior optional (and not active by default to avoid python calling Rust calling python)?

Related to #214

Reproduce

Seen there e.g.: https://github.com/datalayer/jupyter-nbmodel-client/actions/runs/13393090927/job/37405238714

Expected behavior

No error

Context

• version: 0.12.8

Troubleshoot Output

Paste the output from running `jupyter troubleshoot` from the command line here.
You may want to sanitize the paths in the output.

Command Line Output

Paste the output from your command line running `jupyter lab` here, use `--debug` if possible.

Browser Output

Paste the output from your browser Javascript console here, if applicable.

[davidbrochart]

Would it be possible to make that behavior optional

I'm afraid that would make pycrdt not compilable to WASM again.
Could you give a minimal reproducible example?