Add What LLM Can I Run to Open LLM Models

[koraykoylu]

What this adds

What LLM Can I Run? — a free, in-browser hardware
checker, added to Open LLM Models next to the similar can-it-run-llm entry.

Why it fits

• Same purpose as can-it-run-llm, but detects the visitor's hardware live via WebGPU
  (incl. an optional memory-bandwidth benchmark) instead of manual input
• Covers 70+ open-weight models: per-quantization sizes, min RAM, KV-cache-by-context
  and tok/s estimates per GPU class
• Methodology fully published: https://whatsmy.fyi/what-llm-can-i-run/methodology
• Free, no signup, zero data stored (runs entirely client-side)

Formatting matches the surrounding entries. Happy to adjust wording or placement.