Use surface matrix for software cursors

A similar change should probably be applied to hardware cursors, though
more complicated. Also, this doesn't actually fix the issue where the
cursor is too small when over a scale=2 surface. Apparently they don't
set their cursor scales to 2. Seems like a client bug? idk

types/wlr_output.c

@@ -429,8 +429,17 @@ void wlr_output_swap_buffers(struct wlr_output *output) {
 		// with a NULL buffer to hide it
 		if (renderer && texture && texture->valid) {
 			float matrix[16];
-			wlr_texture_get_matrix(texture, &matrix, &output->transform_matrix,
+			if (output->cursor.surface) {
+				float translation[16];
+				wlr_matrix_translate(&translation,
+						output->cursor.x, output->cursor.y, 0);
+				wlr_surface_get_matrix(output->cursor.surface,
+						&matrix, &output->transform_matrix, &translation);
+			} else {
+				wlr_texture_get_matrix(texture,
+						&matrix, &output->transform_matrix,
 						output->cursor.x, output->cursor.y);
+			}
 			wlr_render_with_matrix(renderer, texture, &matrix);
 		}
 	}

types/wlr_surface.c

@@ -656,7 +656,7 @@ void wlr_surface_get_matrix(struct wlr_surface *surface,
 	if (transform) {
 		wlr_matrix_mul(matrix, transform, matrix);
 	}
-	wlr_matrix_scale(&scale, width, height, surface->current->scale);
+	wlr_matrix_scale(&scale, width, height, 1);
 	wlr_matrix_mul(matrix, &scale, matrix);
 	wlr_matrix_mul(projection, matrix, matrix);
 }