LOCA2 First Implementation

[BenGalewsky]

This pull request introduces several key changes to the project, primarily focusing on setting up a CI/CD pipeline, configuring development tools, and adding new functionalities for handling downscaled climate data. Below is a summary of the most important changes:

CI/CD Pipeline Setup:

• Added a GitHub Actions workflow to automate testing, building, and deploying the project. This includes setting up Python, installing dependencies, running flake8 for linting, running pytest for tests, and building and pushing a Docker image. (.github/workflows/ci.yaml)

Development Environment Configuration:

• Added a .dockerignore file to exclude the virtual environment directory. (.dockerignore)
• Configured flake8 to exclude certain directories and set specific linting rules. (.flake8)
• Created a Dockerfile to build a Docker image for the project. (Dockerfile)

Project Documentation:

• Updated the README.md with detailed instructions on setting up the project, configuring access to storage buckets, running the Dagster dashboard, and understanding the project structure. (README.md)

Dagster Integration:

• Defined Dagster assets for downloading raw LOCA2 data and converting it to Zarr format, including configuration for S3 storage. (downscaled_climate_data/assets/loca2.py)
• Set up Dagster definitions for assets, sensors, and resources, integrating them into the project. (downscaled_climate_data/definitions.py)
• Added assets to create intake-esm catalogs for the new datasets

Miscellaneous:

• Added versioning to the project using importlib.metadata. (downscaled_climate_data/__init__.py)
• Configured Dagster to limit the number of concurrent runs to 1 on development deployments. (dagster.yaml)