Add ringkernel-python to CHANGELOG and README for v0.4.0

- Add Python bindings section to CHANGELOG.md with full feature list
- Add Python Bindings section to README.md with installation and usage
- Add ringkernel-python to crate structure table

Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>

Author: mivertowski

CHANGELOG.md

@@ -15,6 +15,58 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 This release extracts ~7,000+ lines of proven GPU infrastructure from RustGraph into RingKernel, making these capabilities available to all RingKernel users.
 
+#### Python Bindings (`ringkernel-python`) - **NEW CRATE**
+
+- **PyO3-based Python wrapper** providing Pythonic access to RingKernel
+  - Full async/await support with `pyo3-async-runtimes` and tokio integration
+  - Sync fallbacks for all async operations (`create_sync`, `launch_sync`, etc.)
+  - Type stubs (`.pyi` files) for IDE support and static type checking
+  - Python 3.8+ compatibility via `abi3-py38`
+
+- **Core Runtime API**:
+  - `RingKernel.create()` / `create_sync()` - Create runtime with backend selection
+  - `KernelHandle` - Launch, activate, deactivate, terminate kernels
+  - `LaunchOptions` - Configure queue capacity, block size, priority
+  - `MessageId`, `MessageEnvelope` - Message handling primitives
+  - `HlcTimestamp`, `HlcClock` - Hybrid Logical Clock support
+  - `K2KBroker`, `K2KEndpoint` - Kernel-to-kernel messaging
+  - `QueueStats` - Queue monitoring and statistics
+
+- **CUDA Support** (feature-gated via `cuda`):
+  - `CudaDevice` - Device enumeration and properties
+  - `GpuMemoryPool` - Stratified GPU memory pool management
+  - `StreamManager` - Multi-stream execution management
+  - `ProfilingSession` - GPU profiling and metrics collection
+
+- **Benchmark Framework** (feature-gated via `benchmark`):
+  - `BenchmarkSuite`, `BenchmarkConfig` - Comprehensive benchmarking
+  - `BenchmarkResult` - Results with throughput and timing
+  - Regression detection with baseline comparison
+  - Multiple report formats (Markdown, JSON, LaTeX)
+
+- **Hybrid Dispatcher**:
+  - `HybridDispatcher` - Automatic CPU/GPU workload routing
+  - `HybridConfig`, `ProcessingMode` - Configuration with adaptive thresholds
+  - `HybridStats` - Execution statistics and threshold learning
+
+- **Resource Management**:
+  - `ResourceGuard` - Memory limit enforcement with safety margins
+  - `ReservationGuard` - RAII wrapper for guaranteed allocations
+  - `MemoryEstimate` - Workload memory estimation
+
+```python
+import ringkernel
+import asyncio
+
+async def main():
+    runtime = await ringkernel.RingKernel.create(backend="cpu")
+    kernel = await runtime.launch("processor", ringkernel.LaunchOptions())
+    await kernel.terminate()
+    await runtime.shutdown()
+
+asyncio.run(main())
+```
+
 #### PTX Compilation Cache (`ringkernel-cuda/src/compile/`) - **NEW MODULE**
 
 - **`PtxCache`** - Disk-based PTX compilation cache for faster kernel loading

README.md

@@ -95,6 +95,50 @@ ringkernel completions bash > ~/.bash_completion.d/ringkernel
 
 Templates available: `basic`, `persistent-actor`, `wavesim`, `enterprise`
 
+## Python Bindings
+
+RingKernel provides Python bindings via PyO3 with full async/await support:
+
+```bash
+# Install from PyPI (when published)
+pip install ringkernel
+
+# Or build from source
+cd crates/ringkernel-python
+maturin develop --features cuda,benchmark
+```
+
+```python
+import ringkernel
+import asyncio
+
+async def main():
+    # Create runtime with CPU backend
+    runtime = await ringkernel.RingKernel.create(backend="cpu")
+
+    # Launch a kernel with options
+    options = ringkernel.LaunchOptions(queue_capacity=1024)
+    kernel = await runtime.launch("processor", options)
+
+    # Kernel lifecycle
+    await kernel.deactivate()
+    await kernel.activate()
+    await kernel.terminate()
+
+    await runtime.shutdown()
+
+asyncio.run(main())
+```
+
+**Features:**
+- Async/await with sync fallbacks (`create_sync`, `launch_sync`, etc.)
+- HLC timestamps and K2K messaging
+- CUDA device enumeration and GPU memory pool management
+- Benchmark framework with regression detection
+- Hybrid CPU/GPU dispatcher with adaptive thresholds
+- Resource guard for memory limit enforcement
+- Type stubs for IDE support (`.pyi` files)
+
 ## Enterprise Runtime
 
 For production deployments, use the enterprise runtime with built-in health checking, circuit breakers, and observability:
@@ -558,6 +602,7 @@ See the [Showcase Applications Guide](docs/15-showcase-applications.md) for deta
 | `ringkernel-cuda-codegen` | Rust-to-CUDA transpiler for writing GPU kernels in Rust DSL |
 | `ringkernel-wgpu-codegen` | Rust-to-WGSL transpiler for writing GPU kernels in Rust DSL (WebGPU) |
 | `ringkernel-ecosystem` | Framework integrations (Actix, Axum, Tower, gRPC) + ML bridges |
+| `ringkernel-python` | Python bindings via PyO3 with async/await support |
 | `ringkernel-wavesim` | 2D wave simulation with tile-based FDTD, ring kernel actors with K2K messaging, and educational modes |
 | `ringkernel-wavesim3d` | 3D acoustic wave simulation with binaural audio, persistent GPU actors (H2K/K2H messaging, K2K halo exchange, cooperative groups), and volumetric rendering |
 | `ringkernel-txmon` | Transaction monitoring showcase with GPU-accelerated fraud detection |