Parallel rendering pipeline, ring vertex buffer, phase timings

Five optimizations measured on the 100k-entity stress scene (Release,
vsync off): 103 FPS baseline -> 297 FPS.

- Sprite submission and vertex building run on all cores above
  Renderer2DOptions.ParallelThreshold (default 8192). Work is sliced
  into 4096-entity segments: a Friflo chunk holds a whole archetype,
  so per-chunk parallelism degenerates to one thread. Segments merge
  in deterministic order, preserving radix sort stability.
- Vertex buffer is ring-written with SetDataOptions.NoOverwrite
  (GPU buffer 2x frame size); Discard only on wrap-around.
- Texture2DRegion precomputes UVs - four float divisions per sprite
  per frame removed.
- Renderer2D exposes per-phase timings (submit/sort/build/upload/draw),
  shown in the sample HUD - all further optimization is data-driven.
- Sample BounceSystem parallelized the same segmented way.

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

tests/MrGameEng.Graphics.Tests/SpriteBatcherTests.cs

@@ -99,6 +99,53 @@ public class SpriteBatcherTests
         }
     }
 
+    [Fact]
+    public void ChunkedSubmit_MergesInChunkOrder_AndCountsCulled()
+    {
+        var batcher = new SpriteBatcher();
+        var key = SpriteSortKey.Make(0, 0f, 0); // одинаковый ключ — порядок задаётся слиянием чанков
+
+        batcher.BeginChunks([3, 2]);
+
+        var writer1 = batcher.GetChunkWriter(1); // чанки могут заполняться в любом порядке (параллельно)
+        writer1.Add(Instance(10), key);
+        writer1.AddCulled();
+        batcher.EndChunk(1, in writer1);
+
+        var writer0 = batcher.GetChunkWriter(0);
+        writer0.Add(Instance(1), key);
+        writer0.Add(Instance(2), key);
+        batcher.EndChunk(0, in writer0);
+
+        var accepted = batcher.CommitChunks();
+
+        Assert.Equal(3, accepted);
+        Assert.Equal(1, batcher.LastChunkCulled);
+        var order = batcher.Sort();
+        Assert.Equal(1, batcher[order[0]].Layer);   // сначала чанк 0...
+        Assert.Equal(2, batcher[order[1]].Layer);
+        Assert.Equal(10, batcher[order[2]].Layer);  // ...затем чанк 1 — стабильно
+    }
+
+    [Fact]
+    public void ChunkedSubmit_GrowsMainArrays_WhenNeeded()
+    {
+        var batcher = new SpriteBatcher(initialCapacity: 2);
+        batcher.BeginChunks([10]);
+
+        var writer = batcher.GetChunkWriter(0);
+        for (byte i = 0; i < 10; i++)
+        {
+            writer.Add(Instance(i), i);
+        }
+
+        batcher.EndChunk(0, in writer);
+
+        Assert.Equal(10, batcher.CommitChunks());
+        Assert.Equal(10, batcher.Count);
+        Assert.Equal(0, batcher[batcher.Sort()[0]].Layer);
+    }
+
     [Fact]
     public void Clear_ResetsCount_KeepsWorking()
     {