Add daemon-attached goal recovery operator

Resume goals after transient API failures with persisted exponential backoff while respecting pauses and interruptions.

codex-keep/store_test.go

@@ -0,0 +1,63 @@
+package main
+
+import (
+	"errors"
+	"os"
+	"path/filepath"
+	"testing"
+	"time"
+)
+
+func closeStore(t *testing.T, s *store) {
+	t.Helper()
+	if err := s.close(); err != nil && !errors.Is(err, os.ErrClosed) {
+		t.Error(err)
+	}
+}
+
+func TestCheckpoint(t *testing.T) {
+	path := filepath.Join(t.TempDir(), "state.json")
+	s, err := openStore(path)
+	if err != nil {
+		t.Fatal(err)
+	}
+	w := failedWatch()
+	w.Due = time.Now().Add(time.Minute).UTC()
+	w.Failures = 3
+	s.Watches["thread-1"] = w
+	if err := s.save(); err != nil {
+		t.Fatal(err)
+	}
+	if other, err := openStore(path); err == nil {
+		closeStore(t, other)
+		t.Fatal("two operators acquired one checkpoint")
+	}
+	closeStore(t, s)
+	restored, err := openStore(path)
+	if err != nil {
+		t.Fatal(err)
+	}
+	defer closeStore(t, restored)
+	r := restored.watch("thread-1")
+	if !r.eligible() || !r.Due.Equal(w.Due) || r.Failures != 3 {
+		t.Fatalf("lost recovery: %+v", r)
+	}
+	info, err := os.Stat(path)
+	if err != nil {
+		t.Fatal(err)
+	}
+	if info.Mode().Perm() != 0600 {
+		t.Fatal(info.Mode())
+	}
+}
+
+func TestBadCheckpoint(t *testing.T) {
+	path := filepath.Join(t.TempDir(), "state.json")
+	if err := os.WriteFile(path, []byte(`{"watches":`), 0600); err != nil {
+		t.Fatal(err)
+	}
+	if s, err := openStore(path); err == nil {
+		closeStore(t, s)
+		t.Fatal("ignored corrupt state")
+	}
+}