fix: call _update_leaf_status in inc/dec_node_lock_ref to prevent phantom evictable nodes

inc_node_lock_ref/dec_node_lock_ref adjusted evictable_size_ but did
not call _update_leaf_status, so nodes could become "phantoms" —
counted in evictable_size_ but missing from evictable_leaves. Under
load with write-behind, this caused eviction to return 0 despite
large evictable_size_, leading to OOM:

  Available tokens: 778624 (evictable_size=706880)
  evict_result=(num_tokens_evicted=0)

The race: write-behind locks node X via inc_node_lock_ref (X stays
in evictable_leaves). A request path touches X via inc_lock_ref,
which calls _update_leaf_status and removes X from evictable_leaves.
Request finishes, dec_lock_ref keeps X out (lock_ref still >0).
Write ack calls dec_node_lock_ref dropping lock_ref to 0, but never
calls _update_leaf_status — X is permanently lost.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

python/sglang/srt/mem_cache/radix_cache.py

@@ -665,6 +665,7 @@ class RadixCache(BasePrefixCache):
             self.evictable_size_ -= len(node.key)
             self.protected_size_ += len(node.key)
         node.lock_ref += 1
+        self._update_leaf_status(node)
 
     def dec_node_lock_ref(self, node: TreeNode):
         """Decrement lock_ref on a single node without walking to root.
@@ -679,6 +680,7 @@ class RadixCache(BasePrefixCache):
             self.evictable_size_ += len(node.key)
             self.protected_size_ -= len(node.key)
         node.lock_ref -= 1
+        self._update_leaf_status(node)
 
     def evictable_size(self):
         return self.evictable_size_