Add StructEval to LLM benchmark suites

[reacher-z]

Summary

• add StructEval to the LLM Benchmark Suites section
• describe its 2,035-example coverage across 18 structured-output formats and 44 generation/conversion task types
• link the official implementation and TMLR paper

Validation

• git diff --check
• verified both added links return HTTP 200
• npx --yes awesome-lint was run; it reports 66 existing README.md baseline errors, with no diagnostic for the changed llm-tools.md

I am a StructEval maintainer and am submitting this entry transparently for maintainer review.

This contribution was prepared with assistance from OpenAI Codex.