Add ONNX2Anything to Model Storage Optimisation

[UnstoppableCurry]

Summary

Adds ONNX2Anything to the Model Storage Optimisation section, right after the ONNX entry — since it's the primary tool for converting ONNX models to deployment-ready formats.

What it does

ONNX2Anything converts ONNX models to edge inference formats (NCNN, MNN, TNN, Tengine, Paddle Lite) entirely in the browser using WebAssembly + Pyodide. No installation, no server, no model uploads. The model never leaves the user's machine.

🌐 Live demo: https://onnx2anything.com
📦 GitHub: https://github.com/UnstoppableCurry/onnx2anything

Fit check

✅ Open source (MIT)
✅ Production-relevant (ONNX is already in the list — this is the primary converter)
✅ Actively maintained
✅ Alphabetical order respected (ONNX → ONNX2Anything → PFA)

[zhimin-z]

That's an interesting one, let's wait for stars and I would merge you in