implement segmentation method for green view score

[banjtheman]

Created a new module to calculate GVI based on the method from the StreetView-NatureVisibility project

README provides steps to install and run locally.

[danbjoseph]

Thanks for working on this! It will be a valuable addition to the project.

• can you please resubmit this against this PR against main? i find it confusing to have some already merged commits showing as changes alongside the changes you've made. the branch you've submitted against is 4 commits behind main.
• the readme instructions should build on our existing readme. for example, we already have a "Setup" section.
• what's the purpose of including a streamlit app?
• this process should be a simple swap for our treepedia based option in 3. Assign a Green View score to each image/feature - when @dragonejt in #42 is tackling having 2 options for step 2, he has gone the route of including an argument for the method. that is, when running python -m src.download_images the user can include a MAPILLARY or LOCAL argument. similarly, for step 3 calculating green view, we may want to let users include an argument defining their preferred analysis method, something like TREEPEDIA or MASK2FORMER

[banjtheman]

• Changed to main branch
• Streamlit app provides a way to visualize how the segmentation works. It is open source and runs locally.
• This part sounds like a new feature, since we would be editing the main script and pipeline. Let's have this PR be for integrating the standalone module.
• We can then can create a new issue to migrate the module into the existing pipeline with the desired functionally (and the README update) (This also provides a good first issue for folks wanting to contribute!)

[dragonejt]

Agree with Jay, here's how I did something similar for image sources in #42 :

• Add a submodule for this step of the project (src.gvi)
• Add an interface (AbstractBaseClass) that both the classical GVI method and new segmentation GVI method will inherit from. Something like src/gvi/gvi_scorer.py, and there should at the very least be a common constructor and a get_gvi_for_image(image_path: Path) abstract method or similar
• Add an Enum to select which GVI Scorer to use, and add it as a CLI argument

[danbjoseph]

this PR addresses #10

[danbjoseph]

it would also be good to save the green score from this segmentation method to a column with a specific header so that the user could run both the "pixel-counting" method and this method on the same files, end up with both result columns, and then compare.

[danbjoseph]

this is superseded by #55