perf(ui): add additive DispatchFused to collapse consecutive UI passes (#47)

Add a fifth UI compute path — DispatchFused — alongside the four
per-element Dispatch* calls. It composites quads → circles → images →
text (canonical back-to-front order) in ONE dispatch that loads the
destination image once and stores once, eliminating the per-pass
load+store and the inter-pass VkMemoryBarriers a run of consecutive
Dispatch* calls would pay. The win materialises at >= 2 non-empty
categories; an absent category is a zero-trip, push-constant-uniform
loop (~free).

The new uber-kernel (shaders/ui-fused.comp.glsl) reuses each category's
exact cooperative shared-memory tile-cull + per-pixel accumulate, so a
fused category is pixel-identical to its standalone pass. Categories run
sequentially per thread and share one set of shared-memory scratch, so
the VGPR/shared high-water mark stays at the per-category level, not the
sum.

Additive and non-breaking: the per-element Dispatch* calls and the
frozen 48-byte UIDispatchHeader are untouched. DispatchFused gets its own
128-byte push-constant layout (UIFusedHeader: four uvec4 headers + four
per-category clip vec4s). To interleave a custom ui.Dispatch() between
standard passes, bracket it with two DispatchFused calls — the dispatch
boundary is the explicit, app-declared flush point.

WebGPU (Vulkan-second): DispatchFused falls back to the per-element
Dispatch* calls in canonical order — one texture/sampler per dispatch
there can't feed the bindless image phase — so the result matches; only
the load/store fusion is Vulkan-only.

HelloUI now composites its background quads, mouse circle, and label
text in a single DispatchFused. Verified rendering identical on a live
Vulkan/Wayland session.

Resolves #47

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

examples/HelloUI/main.cpp

@@ -170,14 +170,20 @@ int main() {
             0, 0, 0, 0,
         };
 
-        // Flush + dispatch. The library inserts the inter-dispatch barriers.
+        // Flush the buffers the GPU is about to read, then composite all three
+        // categories — background quads, the mouse circle, and the button/label
+        // text — in ONE fused dispatch (issue #47). The frame runs these passes
+        // back-to-back with no custom shader interleaved, so DispatchFused loads
+        // and stores the swapchain image once instead of three times and skips
+        // the two inter-pass barriers DispatchQuads→Circles→Text would insert.
+        // Canonical order is quads → circles → images → text; here there are no
+        // images, so that category is a free no-op.
         if (qc > 0) {
 #ifndef CRAFTER_GRAPHICS_WINDOW_DOM
             quadsBuf.FlushDevice(cmd, VK_ACCESS_SHADER_READ_BIT, VK_PIPELINE_STAGE_COMPUTE_SHADER_BIT);
 #else
             quadsBuf.FlushDevice();
 #endif
-            ui.DispatchQuads(cmd, quadsSlot, qc);
         }
         if (cc > 0) {
 #ifndef CRAFTER_GRAPHICS_WINDOW_DOM
@@ -185,7 +191,6 @@ int main() {
 #else
             circlesBuf.FlushDevice();
 #endif
-            ui.DispatchCircles(cmd, circlesSlot, cc);
         }
         if (gc > 0) {
 #ifndef CRAFTER_GRAPHICS_WINDOW_DOM
@@ -193,8 +198,12 @@ int main() {
 #else
             glyphsBuf.FlushDevice();
 #endif
-            ui.DispatchText(cmd, glyphsSlot, gc);
         }
+        ui.DispatchFused(cmd,
+                         {quadsSlot,   qc},
+                         {circlesSlot, cc},
+                         {},                  // no images this frame
+                         {glyphsSlot,  gc});
     });
 
     window.FinishInit();