Fix compute shader so uniforms are the same

Previously,

- set 0 and 1 were read-only, and
- set 1 was write-only

Metal's SPIR-V reflection implementation correctly distinguished these
attributes, and resulted in the failure described by issue
godotengine/godot#101696

A fix for Godot's default SPIR-V reflection is found in godotengine/godot#103614

To ensure the demo keeps working, we ensure all sets have matching
formats.

compute/texture/water_plane/water_compute.glsl

@@ -5,48 +5,48 @@
 layout(local_size_x = 8, local_size_y = 8, local_size_z = 1) in;
 
 // Our textures.
-layout(r32f, set = 0, binding = 0) uniform restrict readonly image2D current_image;
-layout(r32f, set = 1, binding = 0) uniform restrict readonly image2D previous_image;
-layout(r32f, set = 2, binding = 0) uniform restrict writeonly image2D output_image;
+layout(r32f, set = 0, binding = 0) uniform restrict image2D current_image;
+layout(r32f, set = 1, binding = 0) uniform restrict image2D previous_image;
+layout(r32f, set = 2, binding = 0) uniform restrict image2D output_image;
 
 // Our push PushConstant.
 layout(push_constant, std430) uniform Params {
-	vec4 add_wave_point;
-	vec2 texture_size;
-	float damp;
-	float res2;
+    vec4 add_wave_point;
+    vec2 texture_size;
+    float damp;
+    float res2;
 } params;
 
 // The code we want to execute in each invocation.
 void main() {
-	ivec2 tl = ivec2(0, 0);
-	ivec2 size = ivec2(params.texture_size.x - 1, params.texture_size.y - 1);
+    ivec2 tl = ivec2(0, 0);
+    ivec2 size = ivec2(params.texture_size.x - 1, params.texture_size.y - 1);
 
-	ivec2 uv = ivec2(gl_GlobalInvocationID.xy);
+    ivec2 uv = ivec2(gl_GlobalInvocationID.xy);
 
-	// Just in case the texture size is not divisable by 8.
-	if ((uv.x > size.x) || (uv.y > size.y)) {
-		return;
-	}
+    // Just in case the texture size is not divisable by 8.
+    if ((uv.x > size.x) || (uv.y > size.y)) {
+        return;
+    }
 
-	float current_v = imageLoad(current_image, uv).r;
-	float up_v = imageLoad(current_image, clamp(uv - ivec2(0, 1), tl, size)).r;
-	float down_v = imageLoad(current_image, clamp(uv + ivec2(0, 1), tl, size)).r;
-	float left_v = imageLoad(current_image, clamp(uv - ivec2(1, 0), tl, size)).r;
-	float right_v = imageLoad(current_image, clamp(uv + ivec2(1, 0), tl, size)).r;
-	float previous_v = imageLoad(previous_image, uv).r;
+    float current_v = imageLoad(current_image, uv).r;
+    float up_v = imageLoad(current_image, clamp(uv - ivec2(0, 1), tl, size)).r;
+    float down_v = imageLoad(current_image, clamp(uv + ivec2(0, 1), tl, size)).r;
+    float left_v = imageLoad(current_image, clamp(uv - ivec2(1, 0), tl, size)).r;
+    float right_v = imageLoad(current_image, clamp(uv + ivec2(1, 0), tl, size)).r;
+    float previous_v = imageLoad(previous_image, uv).r;
 
-	float new_v = 2.0 * current_v - previous_v + 0.25 * (up_v + down_v + left_v + right_v - 4.0 * current_v);
-	new_v = new_v - (params.damp * new_v * 0.001);
+    float new_v = 2.0 * current_v - previous_v + 0.25 * (up_v + down_v + left_v + right_v - 4.0 * current_v);
+    new_v = new_v - (params.damp * new_v * 0.001);
 
-	if (params.add_wave_point.z > 0.0 && uv.x == floor(params.add_wave_point.x) && uv.y == floor(params.add_wave_point.y)) {
-		new_v = params.add_wave_point.z;
-	}
+    if (params.add_wave_point.z > 0.0 && uv.x == floor(params.add_wave_point.x) && uv.y == floor(params.add_wave_point.y)) {
+        new_v = params.add_wave_point.z;
+    }
 
-	if (new_v < 0.0) {
-		new_v = 0.0;
-	}
-	vec4 result = vec4(new_v, new_v, new_v, 1.0);
+    if (new_v < 0.0) {
+        new_v = 0.0;
+    }
+    vec4 result = vec4(new_v, new_v, new_v, 1.0);
 
-	imageStore(output_image, uv, result);
+    imageStore(output_image, uv, result);
 }