fix(nebulagraph): retry network removal on active endpoints race

After container termination, Docker may not have fully disconnected
endpoints from the network yet. The moby client surfaces this as
"has active endpoints" on NetworkRemove. This race was latent with
the old docker/docker client but is now consistently triggered.

Retry up to 3 times with a 500ms delay when the error indicates
active endpoints remain.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: 2 people

modules/nebulagraph/nebulagraph.go

@@ -4,6 +4,8 @@ import (
 	"context"
 	"errors"
 	"fmt"
+	"strings"
+	"time"
 
 	"github.com/testcontainers/testcontainers-go"
 	"github.com/testcontainers/testcontainers-go/network"
@@ -119,8 +121,19 @@ func terminateContainersAndRemoveNetwork(ctx context.Context, netRes *testcontai
 		}
 	}
 
-	if err := netRes.Remove(ctx); err != nil {
+	// Retry network removal: after container termination, Docker may not
+	// have fully disconnected endpoints yet, causing "has active endpoints".
+	for range 3 {
+		err := netRes.Remove(ctx)
+		if err == nil {
+			break
+		}
+		if strings.Contains(err.Error(), "has active endpoints") {
+			time.Sleep(500 * time.Millisecond)
+			continue
+		}
 		errs = append(errs, fmt.Errorf("network remove: %w", err))
+		break
 	}
 
 	return errs