[v0.1]Bench: Router benchmark CLI

[rootfs]

Acceptance

Command run_bench.sh with:

• Per-category metrics: accuracy, response time, token counts (prompt/completion/total)
• Per-model metrics: success rate, error distribution, latency distribution
• Export to CSV/JSON for analysis

[tao12345666333]

Will this be an independent CLI? You can assign this to me.

[rootfs]

@tao12345666333 the cli is there now (by #125)
https://github.com/vllm-project/semantic-router/tree/main/bench/vllm_semantic_router_bench