Refactor parallel dispatch to use Claude Code agent teams (Wave 4 of #202)

Goal

Refactor /issue and /adversarial-planner to use Claude Code's native agent-team primitives (TeamCreate, TeamDelete, Agent() with team_name) for phases that benefit from parallel dispatch.

Parent: #202 (Workflow optimizations — Waves 1-3 landed in PR #217)

Background

This is Wave 4 / PR-H from the #202 plan. It was intentionally deferred because:

1. Agent teams are experimental (CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1)
2. It's the highest-risk item (touches core orchestrator)
3. It depends on Waves 1-3 landing first (especially PR-E parallel critics and PR-F/G SKILL.md flow changes)

Design decisions (from #202 plan, approved)

1. Teams only for parallel phases. Sequential phases (planner, experimenter, reviewer) keep direct Agent() calls. Teams add value only where 2+ agents run concurrently.
2. Marker comments remain primary state. Teams are ephemeral (die with the session); markers are durable (survive across sessions). No migration of state from markers to team context.
3. Teams created per invocation, not persistent. Since teams don't survive session boundaries, each /issue <N> invocation creates a fresh team if needed for the current phase.
4. isolation: "worktree" replaces manual git worktree add for implementer dispatch in Step 4a.

Phases that benefit from teams

Fallback design (required)

If the teams API is too limited (idle/wake management too complex, limitations block us):

• Fall back to 3 Agent() calls with run_in_background: true
• Feature flag: USE_AGENT_TEAMS=true in the skill so it can be reverted without touching the rest of the flow

Known API limitations (from docs)

• No session resumption with in-process teammates
• Task status can lag
• No nested teams
• One team per session
• Lead is fixed
• Teammates go idle after every turn (need explicit SendMessage to wake)

Acceptance criteria

• /adversarial-planner Phase 2 runs 3 critics via a team (or fallback) on a test task
• /issue Step 4a uses isolation: "worktree" instead of manual git worktree add
• Feature flag allows reverting to non-team dispatch
• Dry-run on a synthetic issue completes full lifecycle
• All existing tests pass
• Documented in SKILL.md with the new flow

Compute

0 GPU-hours. Pure infra work.

Estimate

8-12 hours (multi-day). Develop in a worktree.