Add CONTRIBUTING.md with setup and contribution guidelines

CONTRIBUTING.md

@@ -0,0 +1,86 @@
+# Contributing to neural-lam
+
+Thank you for your interest in contributing to neural-lam! This document
+provides guidelines to help you get started.
+
+## Setting Up the Development Environment
+
+1. Fork the repository on GitHub
+2. Clone your fork locally:
+```bash
+   git clone https://github.com/YOUR_USERNAME/neural-lam.git
+   cd neural-lam
+```
+3. Install dependencies using `uv`:
+```bash
+   uv sync --group dev
+```
+4. Install pre-commit hooks:
+```bash
+   pre-commit install
+```
+
+## Running Tests
+
+Run the full test suite locally before submitting a PR:
+```bash
+pytest tests/
+```
+
+## Running Pre-commit Hooks
+
+To check your code before committing:
+```bash
+pre-commit run --all-files
+```
+
+This will run `black`, `isort`, `flake8`, `mypy` and `codespell` on your code.
+
+## Branch Naming Conventions
+
+Please use the following prefixes for your branches:
+
+| Prefix | Use for |
+|--------|---------|
+| `feat/` | New features or enhancements |
+| `fix/` | Bug fixes |
+| `docs/` | Documentation changes |
+| `test/` | Adding or updating tests |
+| `maintenance/` | CI/CD, dependencies, tooling |
+
+Example: `feat/add-type-hints-metrics`
+
+## Pull Request Checklist
+
+Before opening a PR, make sure you have:
+
+- [ ] Branched off from the latest `main`
+- [ ] Written or updated tests for your changes
+- [ ] Added or updated docstrings for any new/modified functions
+- [ ] Run `pre-commit run --all-files` with no errors
+- [ ] Added an entry to `CHANGELOG.md`
+- [ ] Linked the relevant issue in the PR description
+
+## Commit Message Style
+
+Use the imperative verb form for commit messages:
+
+- ✅ `Add type hints to metrics.py`
+- ✅ `Fix missing mask in loss computation`
+- ❌ `Added type hints`
+- ❌ `Fixing bug`
+
+## Code Style
+
+This project uses:
+- `black` for formatting (line length: 80)
+- `isort` for import sorting
+- `flake8` for linting
+- `mypy` for type checking
+
+All of these are enforced via pre-commit hooks.
+
+## Getting Help
+
+- Open a [GitHub Issue](https://github.com/mllam/neural-lam/issues)
+- Join the MLLAM Slack: https://kutt.it/mllam