feat(#1866): agent-side self-load of configured agent_skills

Each of the 22 consumer agents now self-loads its configured agent_skills
in its mandatory init step, so .planning/config.json agent_skills.<type>
reaches the agent on every runtime — including Cursor and /gsd-autonomous,
where Skill()-delegated workflow bash init did not reliably execute.

- gsd-core/references/agent-skills-bootstrap.md: shared contract
  (query + Read + dedup guard that skips when <agent_skills> is already
  in the prompt, so Claude's orchestrator-side injection never doubles)
- 22 agents/gsd-*.md: one self-load line naming the agent's own type
- gsd-core/workflows/autonomous.md: note that delegated agents self-load
- tests/agent-skills-bootstrap.test.cjs: regression + parity (CONSUMER_AGENTS
  bijection + fast-check property) — Generative-Fix-Divergence guard
- docs: ADR-1866, CONFIGURATION dual-injection How It Works, INVENTORY
  row, Changed changeset

Closes #1866

gsd-core/workflows/autonomous.md

@@ -114,6 +114,8 @@ If `TO_PHASE` is set, display: `Stopping after phase ${TO_PHASE}`
 If `INTERACTIVE` is set, display: `Mode: Interactive (discuss inline, plan+execute inline — background on Codex only)`
 If `PLAN_STRATEGY` is `converge`, display: `Planning: Plan-review convergence enabled`
 
+**Agent skills (delegated agents self-load):** This workflow delegates plan/execute/review via flat `Skill()` invocations rather than resolving `agent_skills` itself. Each consumer agent (`gsd-planner`, `gsd-executor`, `gsd-plan-checker`, `gsd-verifier`, …) self-loads its configured `.planning/config.json` `agent_skills` in its own mandatory init step per `@~/.claude/gsd-core/references/agent-skills-bootstrap.md`. This is the durable path that works on every runtime — including Cursor, where `Skill()`-delegated workflow bash init does not reliably execute. No per-delegation injection is needed here. See open-gsd/gsd-core#1866.
+
 </step>
 
 <step name="discover_phases">