refactor: register CUDA RoPE kernels and fusion in sd.cpp

src/core/ggml_extensions.h

@@ -6,6 +6,7 @@
 void sd_ggml_register_extensions(ggml_backend_t backend);
 
 #ifndef SD_USE_UPSTREAM_GGML
+const struct ggml_ext_op* sd_ggml_rope_op();
 ggml_tensor* sd_ggml_rope_apply(ggml_context* ctx, ggml_backend_t backend, ggml_tensor* x, ggml_tensor* pe, bool interleaved);
 #endif