e2e: allow load generator to succed for short tests (#6952)

This should address last night's failure. We've taken the perspective
of "the load generator shouldn't cause tests to fail" in recent
days/weeks, and I think this is just a next step along that line. The
e2e tests shouldn't test performance. 

I included some comments indicating the ways that this isn't ideal (it
is perhaps not), and I think that if test networks could make
assertions about the required rate, that might be a cool future
improvement (and good, perhaps, for system benchmarking.)

test/e2e/runner/load.go

@@ -54,9 +54,25 @@ func Load(ctx context.Context, testnet *e2e.Testnet) error {
 		case numSeen := <-chSuccess:
 			success += numSeen
 		case <-ctx.Done():
-			if success == 0 {
+			// if we couldn't submit any transactions,
+			// that's probably a problem and the test
+			// should error; however, for very short tests
+			// we shouldn't abort.
+			//
+			// The 2s cut off, is a rough guess based on
+			// the expected value of
+			// loadGenerateWaitTime. If the implementation
+			// of that function changes, then this might
+			// also need to change without more
+			// refactoring.
+			if success == 0 && time.Since(started) > 2*time.Second {
 				return errors.New("failed to submit any transactions")
 			}
+
+			// TODO perhaps allow test networks to
+			// declare required transaction rates, which
+			// might allow us to avoid the special case
+			// around 0 txs above.
 			rate := float64(success) / time.Since(started).Seconds()
 
 			logger.Info("ending transaction load",