Potential performance issue: .to_json memory leak in pandas below 1.4 version

**Issue Description:**

Hello.
I have discovered a performance degradation in the `.to_json` function of pandas version 1.1.0. And I notice the repository depends on pandas 1.1.0 in `docker/lab/files/requirements.txt`. I am not sure whether this performance problem in pandas will affect this repository. I found some discussions on pandas GitHub related to this issue, including [#43877](https://github.com/pandas-dev/pandas/issues/43877) and [#45489](https://github.com/pandas-dev/pandas/pull/45489).
I also found that `ai/metalearning/export_to_mongo.py` used the influenced api. There may be more files using the influenced api.

**Suggestion**

I would recommend considering an upgrade to a different version of pandas >= 1.4 or exploring other solutions to optimize the performance of `.to_json`.
Any other workarounds or solutions would be greatly appreciated.
Thank you!