show-benchmarklisted

# Show a Benchmark Every authoring mission in this repo closes with a `show-*` synthesis instead of dumping its stored deliverable into the conversation (`show-otel-instrumentation-report`, `show-observe-run-report`) - authoring a k6 benchmark is no exception. ## What to render - **Stored path** - where `create-update-benchmark` wrote the script and manifest (`.odd/benchmarks/<name>/`). - **What it exercises** - target service, the endpoints/operations in scope, the test type (smoke/load/stress/soak/spike/breakpoint). - **Validation** - what the manifest records: `k6 inspect` passed (k6 version, date) and the smoke's result - passed (local or remote target, the URL only when the manifest stores it), declined, not applicable (with the scenarios it could not reach), or the functions it did not cover - and the threshold cross-check: each threshold, the service-side floor it was checked against or none found, and the outcome (reachable, kept with the floor acknowledged, or the value the caller changed it to). One or two lines; a benchmark whose manifest records no validation is an upstream contract failure to surface, not a line to invent. - **Next recommended action** - how to actually run it: `/odd-observe run .odd/benchmarks/<name>/` (drive mode with that benchmark, see `docs/guide/benchmarks.md`). - **For an update**: a short headline of what changed against the previous version - the full diff already lives in the commit, this is the human-re