Make MultiResult docs more practical wrt save_result

(instead of keeping it too abstract)

Author: soxofaan

docs/datacube_construction.rst

@@ -223,9 +223,9 @@ in a single entity that can be used to create or run batch jobs. For example:
 
     from openeo import MultiResult
 
-    cube1 = ...
-    cube2 = ...
-    multi_result = MultiResult([cube1, cube2])
+    res1 = cube1.save_result(...)
+    res2 = cube2.save_result(...)
+    multi_result = MultiResult([res1, res2])
     job = multi_result.create_job()
 
 
@@ -237,9 +237,9 @@ which will be automatically grouped as a multi-result:
 
 .. code-block:: python
 
-    cube1 = ...
-    cube2 = ...
-    job = connection.create_job([cube1, cube2])
+    res1 = cube1.save_result(...)
+    res2 = cube2.save_result(...)
+    job = connection.create_job([res1, res2])
 
 
 .. important::

openeo/rest/multiresult.py

@@ -24,9 +24,9 @@ class MultiResult(FlatGraphableMixin):
 
         from openeo import MultiResult
 
-        cube1 = ...
-        cube2 = ...
-        multi_result = MultiResult([cube1, cube2])
+        res1 = cube1.save_result(...)
+        res2 = cube2.save_result(...)
+        multi_result = MultiResult([res1, res2])
         job = multi_result.create_job()
 
     .. seealso::