Function std::mps_matmul_abt_f32_persistent

int mps_matmul_abt_f32_persistent(
    struct MpsBuf bufDC,
    struct MpsBuf bufB,
    &float out,
    unsigned long M,
    unsigned long N,
    unsigned long K,
)

Same as mps_matmul_abt_f32_chained, but 'b' is ALSO a pre-existing device buffer (typically MpsBuf::upload, e.g. a weight matrix that stays GPU-resident across a whole training run) instead of a plain CPU array. Needed for dH = dY . W2^T when W2 is never read back to a CPU array between steps (see mps_sgd_update_f32_chained below).