Fix: Correctly Pass GH_TOKEN to Application in Docker

[WzDTj]

This pull request resolves a critical issue where the GH_TOKEN was not being correctly parsed when the application runs within a Docker container.

The previous Dockerfile used the exec form of the CMD instruction to start the application, which looked like this:

CMD ["start", "-g", "$GH_TOKEN"]

The reason this approach fails is fundamental to how Docker operates: the exec form of CMD does not run the command inside a command shell.

The value passed to the application for the -g flag was not the actual content of the GH_TOKEN variable. Instead, the application received the literal, un-expanded string $GH_TOKEN, which is not a valid token and therefore caused authentication to fail.