Update dashboard environment lock file

[EZoni]

Overview

Dashboard environment update extracted from #439.

Details

I recreated the environment lock file without changing the base file, so this only updates the existing packages:

cd dashboard
git rm environment-lock.yml
conda activate base
conda-lock --file environment.yml --lockfile environment-lock.yml
git add environment-lock.yml

I propose that we merge this first and then extract from #439 the addition of the AmSC Python client to the environment (base file and lock file) in a separate PR (#456), to keep things cleaner and easier to debug, revert, bisect, etc.

[RemiLehe]

I confirm that the new environment works well on MacOS.

Note that I am getting this warning when downloading the model:

2026/06/18 07:12:54 WARNING mlflow.utils.requirements_utils: Detected one or more mismatches between the model's dependencies and the current Python environment:
 - mlflow (current: 3.13.0, required: mlflow==3.9.0)
 - numpy (current: 2.4.6, required: numpy==2.4.2)
 - pydantic (current: 2.13.4, required: pydantic==2.12.5)
 - scipy (current: 1.17.1, required: scipy==1.17.0)
To fix the mismatches, call `mlflow.pyfunc.get_model_dependencies(model_uri)` to fetch the model's environment and install dependencies using the resulting environment file.
2026/06/18 07:12:54 WARNING mlflow.pyfunc: The version of Python that the model was saved in, `Python 3.11.14`, differs from the version of Python that is currently running, `Python 3.12.13`, and may be incompatible

However, I think that this is a minor issue, and could be fixed later, e.g. by updating the environment for training (and building a new Docker image to push to the NERSC registry).

[EZoni]

I confirm that the new environment works well on MacOS.

Note that I am getting this warning when downloading the model:

2026/06/18 07:12:54 WARNING mlflow.utils.requirements_utils: Detected one or more mismatches between the model's dependencies and the current Python environment:
 - mlflow (current: 3.13.0, required: mlflow==3.9.0)
 - numpy (current: 2.4.6, required: numpy==2.4.2)
 - pydantic (current: 2.13.4, required: pydantic==2.12.5)
 - scipy (current: 1.17.1, required: scipy==1.17.0)
To fix the mismatches, call `mlflow.pyfunc.get_model_dependencies(model_uri)` to fetch the model's environment and install dependencies using the resulting environment file.
2026/06/18 07:12:54 WARNING mlflow.pyfunc: The version of Python that the model was saved in, `Python 3.11.14`, differs from the version of Python that is currently running, `Python 3.12.13`, and may be incompatible

However, I think that this is a minor issue, and could be fixed later, e.g. by updating the environment for training (and building a new Docker image to push to the NERSC registry).

Yes, I have been getting that warning for a long time, well before this PR. I agree we should update the ML training environment too.