Align benchmarks with hyperlight-core and the benchmarks repo

Benchmarks already exist. hluk bench runs cold, cold-snap, warm-restore, warm-stateful, and parallel modes; benchmarks/limits.json sets per-runtime, per-OS ceilings that gate CI; and results are charted per host OS via github-action-benchmark on gh-pages. The remaining work is deeper alignment with how hyperlight-core measures performance and with the shared benchmarks repo, so numbers are comparable across projects (methodology, metric names, and reporting).

Start with the existing hluk bench modes and benchmarks/limits.json, then compare their methodology and metric names with hyperlight-core and the shared benchmarks repo. Review how results are reported through github-action-benchmark on gh-pages. Done means the benchmark methods, names, ceilings, and reporting are aligned well enough for comparable cross-project numbers.