Continuous benchmarking

[stuebingerb]

Performance is certainly a concern for our project. However, there is currently little information and comparability, and with new features we can only manually ensure that we're not slowing things down significantly.

We should have a) more performance benchmarks, and b) have them run continuously, while tracking results to be able to spot at least major changes (with public runners we'll likely not have a fully stable environment for comparison).

[stuebingerb]

Ideas (to be extended):

• https://github.com/benchmark-action/github-action-benchmark
• https://benchmark-action.github.io/github-action-benchmark/dev/bench/