Add support for using run_model_on_task simply

[m7142yosuke]

Reference Issue

Addresses #867

What does this PR implement/fix? Explain your changes.

Allow to sepecify task_id(int or str) in run_model_on_task directly.
(i.e. you can use run_model_on_task(clf, task_id) and run_model_on_task(clf, task))

How should this PR be tested?

Compare the results of

from sklearn import ensemble
from openml import tasks, runs

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

and

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()

[mfeurer]

Thanks a lot for your contribution, very much appreciated!

[codecov-io]

Codecov Report

Merging #888 into develop will increase coverage by 0.01%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           develop    #888      +/-   ##
==========================================
+ Coverage    88.49%   88.5%   +0.01%     
==========================================
  Files           37      37              
  Lines         4319    4324       +5     
==========================================
+ Hits          3822    3827       +5     
  Misses         497     497

Impacted Files	Coverage Δ
openml/runs/functions.py	83.1% <100%> (+0.23%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2b7e740...39d6baa. Read the comment docs.