CUDA kernel fusion reduces kernel launches by 96 per batch on Qwen3.8-27B

The official source reports this update: b11177: CUDA: fuse RMS_NORM + SCALE into one kernel. CUDA: fuse RMS_NORM + SCALE into one kernel ( #29393 ) #28068 builds the GDN q/k l2norm as ggml_scale(ggml_rms_norm(x, eps/n), 1/sqrt(n)). This adds 2 SCALE nodes per GDN layer, 96 extra kernel launches per ubatch on Qwen3.8-27B (48 GDN layers). The extra kernels take no measurable GPU time, but each launch has a host/driver cost. It is small with plain batch processing and about 10x larger with draft-mtp speculative decoding. rms_norm_f32 gets a do_scale flag, the same pattern as do_multiply/do_add, so the fused path shares the kernel, the reduction and the launcher.