auto-round

459aab03 - fix fp_layers issue and force to FP16 on cuda for autoround format inference (#326)

fix fp_layers issue and force to FP16 on cuda for autoround format inference (#326) * fix merge error * fix fp_layers issues * Loosen the restrictions of lm-eval * fix and add ut * fix * API usage does not support fuzzy match * bugfix of UT Signed-off-by: Zhang, Weiwei1 <weiwei1.zhang@intel.com> --------- Signed-off-by: Zhang, Weiwei1 <weiwei1.zhang@intel.com> Co-authored-by: WeiweiZhang1 <weiwei1.zhang@intel.com>