Refactor context

[Taucher2003]

The context is currently just a place to put data. It is not really ordered.

Goal should be to split this into some classes, to better organize the data.
A good idea is probably to use these three categories:

• configuration
• runtime
• result

The configuration area will be filled with the configurations of the user, like gitlab project id or access token.
The runtime area will be used for data that is created at and used at runtime, like the handle of the runner container or pipeline.
The result area is used for data that is created at runtime and will be used at the end to create a summary of the run.