use `uv` instead of `venv`

[arthur-shaw]

From what (little) I understand, uv is better in a few dimensions:

• Captures exact packages used for the project
• Activates virtual environment automatically
• Installs packages locally to the project in a faster way that pip

In effect, it seems like an renv for Python.

[arthur-shaw]

Since the project now requires a Python version in a certain range, one can use uv to fulfill those system requirements even (especially) if one has a different version of Python installed globally on one's system.

In case this helps others (or future me), here's how I got things running with uv:

# create a virtual environment with a specific version of Python
uv venv --python 3.11

# pin that version for the project
uv python pin 3.11

# activate the virtual environment
# for Powershell on Windows
.venv/Scripts/activate

# install the requirements, importantly, having `uv` control which Python version does this
uv pip install -r requirement.txt

# run rissk as usual
python main.py export_path="where/my/files/live" output_file="output/goes/here.csv" survey_version=all

Please feel free to add to this if there is a different/better way. This is what my AI overlord suggested after I skimmed the uv docs too quickly and got stuck.