Automate version management with setuptools-scm and simplify release workflow

[Copilot]

Eliminates manual version file editing and reduces release process from 5 steps to 1. Follows implementation pattern from gallantlab/autoflatten.

Changes

Version Management

• Version numbers now automatically derived from git tags via setuptools-scm
• cortex/version.py imports from auto-generated _version.py with fallback for dev environments
• Removed manual version parsing logic from setup.py
• Development versions auto-increment as X.Y.Z.devN between releases

Release Workflow

• Replaced tag-based workflow with release event trigger
• Builds artifacts in isolated job, publishes to both PyPI and TestPyPI
• Uses trusted publishing (OIDC) instead of API tokens
• Added workflow_dispatch for manual TestPyPI testing
• fetch-depth: 0 ensures setuptools-scm can access full git history

Release Process (see updated RELEASE.md)

# Before: Edit version.py, commit, tag, push, create release, revert version
# After: Create GitHub release with version tag (e.g., v1.3.0)

Prerequisites for Deployment

Repository admins must configure PyPI/TestPyPI environments with trusted publishing before workflow can publish packages.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Simplify release workflow</issue_title>
<issue_description>The current release workflow requires too many steps.
Follow what I implemented in https://github.com/gallantlab/autoflatten:

1. Use setuptools-scm to automatically derive version numbers from git tags
2. Set up a workflow to release to pypi and testpypi upon new github releases.

</issue_description>

<agent_instructions>Make sure to check how I implemented this in https://github.com/gallantlab/autoflatten. I will need to enable environments in pypi/testpypi. But you can assume that they will be available. </agent_instructions>

Comments on the Issue (you are @copilot in this section)

• Fixes Simplify release workflow #586

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.