feat(bench): serving benchmark — the baseline instrument for the migration

A scripted 8-turn conversation through the real pipeline (tool call,
memory recall, librarian route, small talk), run against any base URL so
the same instrument measures Ollama today and llama-server after the
cutover. Non-stream mode records wall time and usage; --stream records
time-to-first-token; a VRAM poller and an optional interleaved second
tenant capture the contention the serving plan calls out.

Planted facts carry a BENCH-MARKER prefix so anything the biographer
learns from a bench run stays recognizable in the memory stores.

First smoke against production: 9.7s mean per simple turn, and TTFT
equals wall time — the client sees nothing until the whole
steward-orchestrate-synthesize pipeline has finished. Time-to-first-word
for desklock is currently the full pipeline latency, which makes
streamed synthesis a first-class goal of the serving work.

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

CHANGELOG.md

@@ -7,6 +7,13 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 
+### Added
+
+- `make bench-serving` — serving-layer benchmark replaying a scripted
+  conversation through the pipeline; measures per-turn wall time, TTFT
+  (`--stream`), token usage and VRAM peak, with an optional interleaved
+  second tenant. Baseline instrument for the serving migration.
+
 ### Changed
 
 - `make setup` now ends with a `pytest --collect-only` pass so a broken environment