Simplify run_model_on_task

[mfeurer]

Currently, we require the user to create a task object:

from sklearn import ensemble
from openml import tasks, runs

clf = ensemble.RandomForestClassifier()
task = tasks.get_task(3954)
run = runs.run_model_on_task(clf, task)
run.publish()

which is actually unnecessary and should be possible like this, too:

from sklearn import ensemble
from openml import tasks, runs

clf = ensemble.RandomForestClassifier()
run = runs.run_model_on_task(clf, 3954)
run.publish()