fix(llm): double the per-client token budget to 100k/min

A tool-calling agent re-sends the whole tool result on every iteration of its
loop, so a single question against a tool returning a full collection burns the
old budget in three turns and answers 429 in the middle of reasoning. Pagination
on those tools is the real fix; this buys room until then and stays well inside
the 500k global ceiling.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>

dev/proxy-kustomization.yaml

@@ -40,7 +40,13 @@ spec:
       VECTORDB_HOST: llm-dev-vectordb
       VECTORDB_PORT: "6379"
       RATE_LIMIT_GLOBAL_TOKENS_PER_MINUTE: "500000"
-      RATE_LIMIT_PER_CLIENT_TOKENS_PER_MINUTE: "50000"
+      # Doubled from 50000. A tool-calling agent re-sends the whole tool
+      # result on every iteration of its loop, so one question against a
+      # tool that returns a full collection (list-products is ~23 KB) can
+      # burn the old budget in three turns and answer 429 mid-reasoning.
+      # Pagination on those tools is the real fix; this buys room until
+      # then, and stays well inside the global ceiling.
+      RATE_LIMIT_PER_CLIENT_TOKENS_PER_MINUTE: "100000"
       # Looked up after consumer.yaml's llm-anonymous KongConsumer first
       # reconciled — config.anonymous needs the real Consumer ID, not the
       # username (Konnect stores it as a literal string otherwise, silently