save
Persists a complete benchmark summary: the experiment row, its problems, solver cases and their captured parameters, every cell run with its per-constraint and per-response detail, confirmation finalists, verification estimates, and (when captured) iteration traces and solver state keyed by run id.
Routed through this database's own BenchmarkResultSink implementation, so a summary saved in one call and an experiment streamed problem-by-problem travel the same code and cannot drift apart. A summary is by definition finished, so it never resumes into an unfinished record of the same name — it always gets a fresh experiment id.
Return
the experiment id assigned to this summary within the database
Parameters
the summary returned by a benchmark experiment's run()
an optional KSL version string recorded with the experiment