Add semantic-cache module: CallAdvisor over a Redis Stack vector index with a real in-process embedding model; threshold sweep (hit rate against wrong answers), 600-request replay with an illustrative cost model, tenant filter, expiry, look-alike and embedding-model-change traps; optional real local model

Co-Authored-By: Claude Sonnet 5.5 <[email protected]>
Claude-Session: https://claude.ai/code/session_01G8ikz8xdWuTP5yun8DZ1hk

semantic-cache/output/04-replay-600-requests.txt

@@ -0,0 +1,16 @@
+# 600 requests through the advisor, by threshold
+
+Workload: 600 requests (seed 42), 30 intents asked with Zipf popularity in random phrasings, 15% off-topic.
+  121 distinct texts, so a plain exact-match cache could answer at most 479 of 600 (80%).
+Cost model (ILLUSTRATIVE): $2.50 per million input tokens, $10.00 per million output tokens, tokens = characters / 4.
+
+setup                  | model calls | hit rate |  wrong | wrong of hits |   model $ | saved
+no cache               |        600 |        - |      - |             - |    0.4564 | -
+cache, threshold 0.70  |         42 |    93.0% |    191 |         34.2% |    0.0316 | 93%
+cache, threshold 0.80  |         62 |    89.7% |      7 |          1.3% |    0.0468 | 90%
+cache, threshold 0.90  |        109 |    81.8% |      0 |          0.0% |    0.0827 | 82%
+cache, threshold 0.95  |        119 |    80.2% |      0 |          0.0% |    0.0904 | 80%
+cache, exact match only |        121 |    79.8% |      0 |          0.0% |    0.0919 | 80%
+
+Average time to serve a cache hit (embed the question, search Redis, build the response): 4 ms at 0.70, 4 ms at 0.80, 3 ms at 0.90, 3 ms at 0.95
+Saved dollars exclude the cost of embedding (local model here, CPU only) and of running Redis.