roc_auc_score fails if dtype is object

to reproduce:

``` python
import numpy as np
np.random.seed(13)
classes = np.array(['yes', 'no'])
y = classes[np.random.randint(2, size=10)]
t = classes[np.random.randint(2, size=10)]
from sklearn.metrics import roc_auc_score
from sklearn.metrics import accuracy_score
roc_auc_score(t, y)
```

```
ValueError: Data is not binary and pos_label is not specified
```
