Fix _Float16 casting issue and reduce LMUL for certain vector instruction from m2 to m1.#5451

martin-frbg merged 1 commit intoOpenMathLib:developOpenMathLib/OpenBLAS:developfrom

ChipKerchner:fixRVVSHGEMMChipKerchner/OpenBLAS:fixRVVSHGEMMCopy head branch name to clipboard