Add support for missing quants in CPY (Metal & CUDA).#11987

Closed

gcp wants to merge 2 commits intoggml-org:masterfrom gcp:quant_cpy_cuda

+470-5

Commits on Feb 19, 2025

cuda: Add Q5_1, Q5_0, Q4_1 and Q4_0 to F32 conversion support. (ggml-org#10976 )
gcp
committed

Commits on Feb 20, 2025

metal: Copy kernels for quant to F32 conversions (ggml-org#10976 ).
gcp
committed