Adds an example showing a generator agent producing code, then three reviewer agents (security, performance, style) analyzing it in parallel, followed by a synthesizer merging all feedback into a prioritized action item report. Closes #75
Matt Van Horn