[EVAL] Add SLR-Bench

## Evaluation short description
- Why is this evaluation interesting?
Scalable Logical Reasoning Benchmark contains 19k+ prompts spanning 20 curriculum levels
Systematic progression, from simple attribute checks to complex recursion
Ideal for both model evaluation and curriculum-based model training
While models readily solve the initial levels, performance drops as complexity increases
- How used is it in the community?
To evaluate reasoning abilities of today’s Large Langauge Models

## Evaluation metadata
Provide all available
- Paper url: https://arxiv.org/abs/2506.15787
- Github url: https://github.com/ml-research/ScalableLogicalReasoning
- Dataset url: https://huggingface.co/datasets/AIML-TUDA/SLR-Bench