fix(graphics): Make step_size a float in the vertex shader

Some OpenGL implementations are stricter than others it seems. On my
Mac, the previous shader happily compiles. On my Ubuntu machine, GL
complains that the same uniform has a different data type in the two
shaders.

Fixes Trac#16772

Author: zjp

src/bundles/graphics/src/vertexShader.txt

@@ -47,7 +47,7 @@ uniform mat4 view_matrix; // Map scene coordinates to camera coordinates
 #endif
 
 #if defined(USE_VOLUME_RAYCASTING)
-uniform vec3 step_size;
+uniform float step_size;
 uniform vec3 camera_pos;
 uniform float camera_fov;
 uniform vec2 window_size;