Add traceAI and ai-evaluation to LLM Ops & Observability section

[SuhaniNagpal7]

Summary

• Adds traceAI — OpenTelemetry-native tracing framework for LLM/agent apps (peer of OpenLIT, Phoenix)
• Adds ai-evaluation — LLM eval framework with 50+ metrics, LLM-as-Judge, and guardrail scanners (peer of Promptfoo, Opik)