[Router] Add --worker-queue-limit: stop sending cache-affinity traffic to a queueing worker (#39168)

Co-authored-by: Kangyan Zhou <kangyan.zhou@radixark.ai>
Co-authored-by: Claude Opus 5 (1M context) <noreply@anthropic.com>

experimental/sgl-router/tests/component/policies/bucket_domains.rs

@@ -154,6 +154,7 @@ fn cache_candidate_uses_bucket_metadata_without_extend_range_filtering() {
         worker: cached,
         matched_prefix_tokens: 128,
         uncached_tokens: 128,
+        matched_prefix_blocks: 8,
         candidate_range_id: "global".into(),
         max_pending_prefill_tokens: None,
     };

experimental/sgl-router/tests/proxy/shared_prefill_admission.rs

@@ -124,6 +124,7 @@ impl Policy for CacheCandidatesPolicy {
                 worker: Arc::clone(&self.worker),
                 matched_prefix_tokens: 1,
                 uncached_tokens: 1,
+                matched_prefix_blocks: 1,
                 candidate_range_id: "global".into(),
                 max_pending_prefill_tokens: None,
             }],