feat(role): add versioned role configs + escalation ladder + scheduler (Phase 5)

Two parts:

Part A (fixes a gap from Phase 1): Groq/OpenRouter/OpenAI were documented in
docs/api-keys-setup.md as usable once configured, but nothing actually
constructed runners for them. internal/cli/cloudrunners.go consolidates
anthropic/google/groq/openrouter/openai NativeRunner construction into one
table-driven registerCloudRunners() helper, replacing the two hand-written
per-provider blocks in serve.go/run.go. Groq/OpenRouter/OpenAI reuse
openaicompat (no new adapter code) at SandboxKind: "docker".

Part B: the token-husbanding harness's core routing mechanism.

- internal/role: RoleConfig/Tier/Rung -- a role's system prompt and a
  multi-tier (provider, model) escalation ladder, versioned via config_json.
- storage: new role_configs table (draft/active/retired, UNIQUE(role,
  version)) with transactional activate-retires-prior-active semantics;
  new executions.escalation_rung column.
- task.AgentConfig.Role string -- purely additive; every existing task shape
  (Agent.Role == "") is unaffected, proven by TestPool_Execute_NonRoleTask_Unaffected
  plus the full pre-existing suite passing unchanged.
- executor.Pool.execute(): role-typed tasks with no Agent.Type yet resolve
  tier 0 of their active ladder (round-robin across multi-candidate tiers,
  skipping rate-limited providers, falling back to soonest-clearing) before
  the existing pickAgent/Classifier path runs; SystemPrompt applies to
  Agent.SystemPromptAppend. Already-resolved role tasks (scheduler resubmits)
  get their escalation_rung re-derived read-only via findTierIndex.
- internal/scheduler: polls role-typed FAILED tasks, retries at the same
  rung under MaxRetries or escalates to the next tier's first candidate when
  budget.Accountant.Allow() permits (emitting event.KindEscalated), else
  leaves the task FAILED with a final:true KindEscalated event. An
  in-memory per-execution-ID "handled" set keeps the poll loop convergent.
  Started by `serve` only, config knob [scheduler].poll_interval_seconds.
- internal/api: POST/GET /api/roles/{role}/versions, POST
  /api/roles/{role}/activate -- unauthenticated, matching the existing
  projects/tasks REST endpoints' auth posture (only chatbot MCP, agent MCP,
  and WebSocket are api_token-gated in this codebase today).

Documented as stored-but-not-yet-enforced (CLAUDE.md Design Debt, matching
how task.Priority/RetryConfig are already documented): RoleConfig.Tools/
SandboxKind don't affect dispatch yet; DefaultBudgetUSD is read narrowly as
the scheduler's escalation cost estimate, not enforced at initial dispatch;
scheduler escalation always targets Candidates[0] (no round-robin, unlike
initial-dispatch tier-0 resolution); the scheduler's dedupe is per-process
and resets on restart (idempotent, harmless).

go build/vet/test -race -count=1 all pass, 21 packages.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01V1moSNCJRcP6kykA4tyUSs

1 files changed, 37 insertions, 6 deletions

diff --git a/internal/config/config.go b/internal/config/config.go
index b9454d1..9e065b5 100644
--- a/internal/config/config.go
+++ b/internal/config/config.go
@@ -72,14 +72,26 @@ type RunnersConfig struct {
 	// Anthropic vs. Claude above. Also requires [providers.google].api_key to
 	// be set; Google = true with no configured key has no effect.
 	Google *bool `toml:"google"`
+	// Groq, OpenRouter, and OpenAI gate NativeRunners registered under agent
+	// types "groq"/"openrouter"/"openai", all backed by
+	// internal/provider/openaicompat (they're OpenAI-wire-compatible, so no
+	// dedicated adapter package is needed — see docs/api-keys-setup.md).
+	// Each also requires the matching [providers.<name>].api_key to be set;
+	// true with no configured key has no effect, same as Anthropic/Google.
+	Groq       *bool `toml:"groq"`
+	OpenRouter *bool `toml:"openrouter"`
+	OpenAI     *bool `toml:"openai"`
 }
 
-func (r RunnersConfig) ClaudeEnabled() bool    { return r.Claude == nil || *r.Claude }
-func (r RunnersConfig) GeminiEnabled() bool    { return r.Gemini == nil || *r.Gemini }
-func (r RunnersConfig) LocalEnabled() bool     { return r.Local == nil || *r.Local }
-func (r RunnersConfig) ContainerEnabled() bool { return r.Container == nil || *r.Container }
-func (r RunnersConfig) AnthropicEnabled() bool { return r.Anthropic == nil || *r.Anthropic }
-func (r RunnersConfig) GoogleEnabled() bool    { return r.Google == nil || *r.Google }
+func (r RunnersConfig) ClaudeEnabled() bool     { return r.Claude == nil || *r.Claude }
+func (r RunnersConfig) GeminiEnabled() bool     { return r.Gemini == nil || *r.Gemini }
+func (r RunnersConfig) LocalEnabled() bool      { return r.Local == nil || *r.Local }
+func (r RunnersConfig) ContainerEnabled() bool  { return r.Container == nil || *r.Container }
+func (r RunnersConfig) AnthropicEnabled() bool  { return r.Anthropic == nil || *r.Anthropic }
+func (r RunnersConfig) GoogleEnabled() bool     { return r.Google == nil || *r.Google }
+func (r RunnersConfig) GroqEnabled() bool       { return r.Groq == nil || *r.Groq }
+func (r RunnersConfig) OpenRouterEnabled() bool { return r.OpenRouter == nil || *r.OpenRouter }
+func (r RunnersConfig) OpenAIEnabled() bool     { return r.OpenAI == nil || *r.OpenAI }
 
 // ProviderConfig configures a native (or OpenAI-compatible) LLM provider
 // backend for the multi-provider harness — see docs/api-keys-setup.md. Phase
@@ -131,6 +143,10 @@ type Config struct {
 	LocalModel      LocalModel    `toml:"local_model"`
 	Runners         RunnersConfig `toml:"runners"`
 	Budget          BudgetConfig  `toml:"budget"`
+	// Scheduler configures internal/scheduler.Scheduler, the retry-then-
+	// escalate loop for role-typed tasks. Started only by `serve` (the
+	// one-shot `run` command has no background scheduler).
+	Scheduler SchedulerConfig `toml:"scheduler"`
 	// Providers configures native/OpenAI-compatible LLM backends by name (see
 	// ProviderConfig). Unused by runner construction in Phase 1.
 	Providers map[string]ProviderConfig `toml:"providers"`
@@ -145,6 +161,21 @@ type BudgetConfig struct {
 	Provider5hUSD map[string]float64 `toml:"provider_5h_usd"`
 }
 
+// SchedulerConfig configures internal/scheduler.Scheduler's poll loop.
+type SchedulerConfig struct {
+	// PollIntervalSeconds is how often the scheduler checks for role-typed
+	// FAILED tasks to retry or escalate. Defaults to 30 when zero/unset.
+	PollIntervalSeconds int `toml:"poll_interval_seconds"`
+}
+
+// PollInterval returns the configured poll interval, defaulting to 30s.
+func (c SchedulerConfig) PollInterval() time.Duration {
+	if c.PollIntervalSeconds <= 0 {
+		return 30 * time.Second
+	}
+	return time.Duration(c.PollIntervalSeconds) * time.Second
+}
+
 func Default() (*Config, error) {
 	home, err := os.UserHomeDir()
 	if err != nil {