Create learning without specifing experiment-id

Enable to create learning without specifing experiment_id.

Currently, user must specify experiment_id and model_id
when creating learning-job.
However, meteso can find out experiment_id from model_id.

Change-Id: I0ba179424bc794469824c934ee3cd6988db70991

meteosclient/api/learnings.py

@@ -28,13 +28,12 @@ class LearningManager(base.ResourceManager):
     NotUpdated = base.NotUpdated()
 
     def create(self, display_name=None, display_description=None,
-               experiment_id=None, model_id=None, method=None, args=None):
+               model_id=None, method=None, args=None):
         """Create a Learning."""
 
         data = {
             'display_name': display_name,
             'display_description': display_description,
-            'experiment_id': experiment_id,
             'model_id': model_id,
             'method': method,
             'args': base64.b64encode(str(args)),

sample/json/learning.json

@@ -1,7 +1,6 @@
 {
     "display_name": "example-learning-job",
     "display_description": "This is a sample job",
-    "experiment_id": "<YOUR_EXPERIMENT_ID>",
     "model_id": "<YOUR_MODEL_ID>",
     "method": "predict",
     "args": "<INPUT_DATA>"