Adde guide for visual testing

[LucaMarconato]

FYI, this is the command that I use for act:

echo '\033[0;31mremove -r when testing a new branch\033[0m' && time act test --artifact-server-path ~/temp/act --env CODECOV_TOKEN=secret_token_you_need_to_generate --container-architecture linux/arm64 -r -j test 

Please notice that you have to change arm64 to amd64 if you are not using Apple Silicon.
It may work even without the codecov token, please ask me if you encounter problem and I'll help with that.

[codecov-commenter]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 83.54%. Comparing base (49434a3) to head (46b0172).

Additional details and impacted files

@@           Coverage Diff           @@
##             main     #397   +/-   ##
=======================================
  Coverage   83.54%   83.54%           
=======================================
  Files           8        8           
  Lines        1689     1689           
=======================================
  Hits         1411     1411           
  Misses        278      278           

[LucaMarconato]

This is my PyCharm configuration for working with the Dockerfile in this repo.

[LucaMarconato]

CC @timtreis @Sonja-Stockhaus I made a Dockerfile to ensure reproducibility of the test plots in any local machine. I did this because I was not able to have the tests passing in my MacBook.

I also updated the contribution guide to explain how to update new artifacts.

[LucaMarconato]

@timtreis I made a mistake and pushed the Dockerfile here 54c0355. I didn't update the CHANGELOG since it's not a new feature/bugfix for users. But I could do it you want.

[LucaMarconato]

If the setup via the GUI doesn't work, one can always pre-build the image manually.

The command to build the image in the spatialdata-plot repo:

docker build --build-arg TARGETPLATFORM=linux/arm64 --tag test-spatialdata-plot -f test.Dockerfile .

[LucaMarconato]

Using Docker in PyCharm was possible but not straightforward due to this bug https://youtrack.jetbrains.com/issue/PY-81541/Failed-to-configure-Docker-based-SDK-non-zero-exit-code-1?backToIssues=false. It has now been fixed in PyCharm 2025.3.1 so the config of the Docker image how I show in the screenshot above works again smoothly now.