docs(benchmark): auto-update performance charts and README

Author: namanONcode

README.md

@@ -16,34 +16,12 @@ zThread is designed for maximum throughput and minimal latency. In single-produc
 <!-- BENCHMARK_TABLE_START -->
 | Framework / Mechanism | Throughput (Higher is better) | Average Latency (Lower is better) | Engine Architecture |
 | :--- | :--- | :--- | :--- |
-| **ConcurrentLinkedQueue** | ~20.22 M ops/sec | ~49.5 ns / event | Lock-free queue algorithm |
-| **ArrayBlockingQueue** | ~19.83 M ops/sec | ~50.4 ns / event | ReentrantLock + Condition queues |
-| **LinkedBlockingQueue** | ~12.33 M ops/sec | ~81.1 ns / event | Two-lock queue algorithm |
-| **zThread (Linux FFM / Epoll)** | **~11.72 M ops/sec** | **~85.3 ns / event** | Kernel `epoll` + Lock-free RingBuffer via Panama FFM |
-| **SynchronousQueue** | ~11.04 M ops/sec | ~90.6 ns / event | Dual stack / queue thread handoff |
-| **Netty (NIO EventLoop)** | ~11.01 M ops/sec | ~90.8 ns / event | `Selector` + ConcurrentLinkedQueue dispatch |
-| **Project Reactor (Schedulers)** | ~9.15 M ops/sec | ~109.3 ns / event | RingBuffer-backed Schedulers |
-| **Vert.x (Event Loop)** | ~8.34 M ops/sec | ~119.9 ns / event | Netty-backed event loop dispatch |
-| **Java Virtual Threads (Loom)** | ~4.91 M ops/sec | ~203.7 ns / event | Carrier thread park/unpark overhead |
-| **benchmark** | ~4.34 M ops/sec | ~230.4 ns / event | Unknown architecture |
-| **benchmark** | ~4.04 M ops/sec | ~247.6 ns / event | Unknown architecture |
-| **benchmark** | ~3.96 M ops/sec | ~252.7 ns / event | Unknown architecture |
-| **benchmark** | ~3.76 M ops/sec | ~266.3 ns / event | Unknown architecture |
-| **benchmark** | ~3.69 M ops/sec | ~270.8 ns / event | Unknown architecture |
-| **benchmark** | ~3.68 M ops/sec | ~271.4 ns / event | Unknown architecture |
-| **benchmark** | ~3.62 M ops/sec | ~276.0 ns / event | Unknown architecture |
-| **benchmark** | ~3.61 M ops/sec | ~276.8 ns / event | Unknown architecture |
-| **benchmark** | ~3.27 M ops/sec | ~305.9 ns / event | Unknown architecture |
-| **benchmark** | ~3.02 M ops/sec | ~330.9 ns / event | Unknown architecture |
-| **benchmark** | ~2.98 M ops/sec | ~335.6 ns / event | Unknown architecture |
-| **benchmark** | ~2.89 M ops/sec | ~345.7 ns / event | Unknown architecture |
-| **benchmark** | ~2.88 M ops/sec | ~347.8 ns / event | Unknown architecture |
-| **benchmark** | ~2.81 M ops/sec | ~355.4 ns / event | Unknown architecture |
-| **benchmark** | ~2.71 M ops/sec | ~369.6 ns / event | Unknown architecture |
-| **benchmark** | ~1.34 M ops/sec | ~746.6 ns / event | Unknown architecture |
-| **benchmark** | ~1.06 M ops/sec | ~940.4 ns / event | Unknown architecture |
-| **benchmark** | ~0.75 M ops/sec | ~1340.4 ns / event | Unknown architecture |
-| **benchmark** | ~0.56 M ops/sec | ~1784.6 ns / event | Unknown architecture |
+| **zThread (Linux FFM / Epoll)** | **~9.79 M ops/sec** | **~102.1 ns / event** | Kernel `epoll` + Lock-free RingBuffer via Panama FFM |
+| **Project Reactor (Schedulers)** | ~7.91 M ops/sec | ~126.4 ns / event | RingBuffer-backed Schedulers |
+| **Netty (NIO EventLoop)** | ~6.39 M ops/sec | ~156.5 ns / event | `Selector` + ConcurrentLinkedQueue dispatch |
+| **Vert.x (Event Loop)** | ~6.13 M ops/sec | ~163.1 ns / event | Netty-backed event loop dispatch |
+| **Java Virtual Threads (Loom)** | ~4.48 M ops/sec | ~223.4 ns / event | Carrier thread park/unpark overhead |
+| **SynchronousQueue** | ~1.36 M ops/sec | ~735.2 ns / event | Dual stack / queue thread handoff |
 <!-- BENCHMARK_TABLE_END -->
 
 ## Installation