Q4_2 quantization with rmse-optimized scale and quants

[ikawrakow]

For quantize-stats we get q4_2: rmse 0.00159301, maxerr 0.17480469, 95pct<0.0030, median<0.0012

For 7B perplexity with BLAS enabled we get 6.2038 after 655 chunks.

Quantization is slow (~90 seconds on my Mac for 7B) as not multi-threaded as in PR #896.