MLFlow ResponseError('too many 500 error responses') if try to log to deleted experiment

[millskyle]

Bug description

If one tries to log to a deleted MLFlow experiment, (e.g. you delete the experiment in the MLFlow UI, go on holidays, come back renewed and refreshed and are greeted by your previously working code no longer working), Lightning throws a ResponseError('too many 500 error responses').

I will create a PR with a proposed simple fix.

What version are you seeing the problem on?

v2.5

How to reproduce the bug

Error messages and logs

mlflow.exceptions.MlflowException: API request to http://localhost:5000/api/2.0/mlflow/runs/create failed with exception HTTPConnectionPool(host='localhost', port=5000): Max retries exceeded with url: /api/2.0/mlflow/runs/create (Caused by ResponseError('too many 500 error responses'))

Environment

Current environment

#- PyTorch Lightning Version (e.g., 2.5.0):
#- PyTorch Version (e.g., 2.5):
#- Python version (e.g., 3.12):
#- OS (e.g., Linux):
#- CUDA/cuDNN version:
#- GPU models and configuration:
#- How you installed Lightning(`conda`, `pip`, source):

More info

No response