feat(executor): add DAG auto-cascade-fail + role-typed subtask spawning (Phase 6)

Two prerequisites for safe parallel evaluator fan-out (later phase):

1. Auto-cascade-fail: previously, a failed task's dependents just sat
   PENDING/QUEUED forever (or until something eventually tried to dispatch
   them and discovered the dependency was dead). Pool.cascadeFail now fires
   right after a task lands in a terminal failure state (FAILED/TIMED_OUT/
   CANCELLED/BUDGET_EXCEEDED, from handleRunResult, the budget-gate reject
   path, and the checkDepsReady dependency-failure path), recursively
   cancelling every not-yet-run dependent (PENDING/QUEUED only -- RUNNING and
   terminal states are left alone) with a message referencing the upstream
   failure. A visited-set guards recursion, which turned out to be
   load-bearing rather than defense-in-depth: task creation does not prevent
   dependency cycles anywhere in this codebase.

   Correction to an earlier assumption: internal/executor's
   waitForDependencies is dead code, never called. The live mechanism is
   checkDepsReady, invoked synchronously in execute() with a self-requeue via
   time.AfterFunc. Added a freshness re-check (GetTask, bail if no longer
   QUEUED) at the top of that block so a task cascade-cancelled while sitting
   in the requeue loop stops silently instead of hitting an invalid
   CANCELLED->CANCELLED transition or, worse, still getting dispatched.

2. storeChannel.SpawnSubtask hardcoded Agent.Type: "claude" on every spawned
   child regardless of what role it should play -- a hard blocker for a
   Planner/Builder task spawning role-typed evaluator subtasks. SubtaskSpec
   (internal/agentchannel) gains a Role field; when set, the child task gets
   Agent.Role instead of a hardcoded Type, so Phase 5's role-resolution picks
   provider/model from that role's escalation ladder. spec.Role == "" (every
   existing caller) preserves today's exact behavior byte-for-byte -- proven
   by an explicit regression test, not just new-feature coverage. Threaded
   the new `role` parameter through both spawn_subtask transports: the native
   tool-use loop (internal/agentloop/tools.go) and the MCP tool exposed to
   ContainerRunner-driven claude/gemini agents (internal/executor/agentmcp.go).

go build/vet/test -race -count=1 all pass, full suite (20 packages).

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

1 files changed, 76 insertions, 0 deletions

diff --git a/internal/storage/db_test.go b/internal/storage/db_test.go
index 09bbdfc..b53234a 100644
--- a/internal/storage/db_test.go
+++ b/internal/storage/db_test.go
@@ -1256,4 +1256,80 @@ func TestUpdateProject(t *testing.T) {
 	}
 }
 
+func makeDepTask(id string, dependsOn []string) *task.Task {
+	now := time.Now().UTC()
+	if dependsOn == nil {
+		dependsOn = []string{}
+	}
+	return &task.Task{
+		ID: id, Name: "Task " + id,
+		Agent:     task.AgentConfig{Type: "claude", Instructions: "x"},
+		Priority:  task.PriorityNormal,
+		Retry:     task.RetryConfig{MaxAttempts: 1, Backoff: "linear"},
+		Tags:      []string{},
+		DependsOn: dependsOn,
+		State:     task.StatePending,
+		CreatedAt: now, UpdatedAt: now,
+	}
+}
+
+// TestListDependents verifies ListDependents returns exactly the tasks that
+// directly depend on the given ID, not transitive dependents and not
+// unrelated tasks.
+func TestListDependents(t *testing.T) {
+	db := testDB(t)
+
+	// A has no deps.
+	// B and C both depend directly on A.
+	// D depends on B (transitive on A, not direct).
+	// E has no relation to A at all.
+	a := makeDepTask("ld-a", nil)
+	b := makeDepTask("ld-b", []string{"ld-a"})
+	c := makeDepTask("ld-c", []string{"ld-a"})
+	d := makeDepTask("ld-d", []string{"ld-b"})
+	e := makeDepTask("ld-e", nil)
+
+	for _, tk := range []*task.Task{a, b, c, d, e} {
+		if err := db.CreateTask(tk); err != nil {
+			t.Fatalf("CreateTask(%s): %v", tk.ID, err)
+		}
+	}
+
+	deps, err := db.ListDependents("ld-a")
+	if err != nil {
+		t.Fatalf("ListDependents: %v", err)
+	}
+	if len(deps) != 2 {
+		t.Fatalf("want 2 direct dependents of ld-a, got %d: %+v", len(deps), deps)
+	}
+	got := map[string]bool{}
+	for _, dep := range deps {
+		got[dep.ID] = true
+	}
+	if !got["ld-b"] || !got["ld-c"] {
+		t.Errorf("expected dependents ld-b and ld-c, got %v", got)
+	}
+	if got["ld-d"] || got["ld-e"] {
+		t.Errorf("ListDependents must not include transitive (ld-d) or unrelated (ld-e) tasks: %v", got)
+	}
+
+	// ld-d depends only on ld-b.
+	depsOfB, err := db.ListDependents("ld-b")
+	if err != nil {
+		t.Fatalf("ListDependents(ld-b): %v", err)
+	}
+	if len(depsOfB) != 1 || depsOfB[0].ID != "ld-d" {
+		t.Errorf("want [ld-d] as dependents of ld-b, got %+v", depsOfB)
+	}
+
+	// A task with no dependents returns an empty slice, not an error.
+	depsOfE, err := db.ListDependents("ld-e")
+	if err != nil {
+		t.Fatalf("ListDependents(ld-e): %v", err)
+	}
+	if len(depsOfE) != 0 {
+		t.Errorf("want no dependents of ld-e, got %+v", depsOfE)
+	}
+}
+