Add performance benchmarking for DFG discovery

[martinkuhn94]

We need accurate runtime measurements for all processing steps and overall execution time, including system metrics (CPU/RAM), and benchmarks across multiple dataset sizes to evaluate scaling.

Acceptance criteria:

• Total Runtime from API Request to final output
• Collect system stats (VM/PC) during runs (CPU/RAM)
• Benchmark scenarios for multiple dataset sizes 10%, 20% .... 100%
• Store results in the Github repository
• Create a figure showing the experiments
• Include them in the documentation / Readme