Unreasonable default for validation stream creation

```{python}
benchmark = SplitCifar100(20, shuffle=True)
benchmark = benchmark_with_validation(benchmark, val_size=0.05)
print(benchmark.valid_stream[0].classes_in_this_experience) # Shows 125 samples of the same class
```

The default value to create the validation stream is shuffle=False. This could be reasonable choice if the benchmark dataset is already shuffled (no need to shuffle again in that case). However, when creating it that way, it seems like it's not shuffled.