[KM-668] feat(eval): model tasks_run; substantive = successful analyze (T2)

[NOTICKET] feat(eval): report-readiness eval (dataset + runner + baseline)

[KM-659] fix(eval): make eval a package; run via -m eval.intent.run_eval

[KM-659] feat(eval): optional Langfuse sink (--langfuse)

[KM-659] feat(eval): intent-routing eval runner + metrics

[KM-658] feat(eval): golden intent dataset for the 6-intent router