UCI regression benchmarks

[murphyk]

Extract the 15 "hard" UCI regression benchmarks used in

Q. Wu, C. Wang, J. Langford, P. Mineiro, and M. Rossi, “ChaCha for Online AutoML,” in ICML, Jun. 2021 [Online]. Available: http://arxiv.org/abs/2106.04815

https://github.com/qingyun-wu/FLAML/blob/ChaCha_new_eval/test/vw/resources/benchmarks/final_icml.yaml

Add to our test pipeline.