CI is broken

[astrojuanlu]

#136 made a one character change in the README but there were some test failures.

https://app.circleci.com/pipelines/github/kedro-org/kedro-starters/618/workflows/687e92b1-d6d7-41e1-9348-dc8566bb9c39/jobs/6650

Failing scenarios:
  features/run.feature:26  Run a Kedro project created from pyspark-iris

0 features passed, 1 failed, 0 skipped
4 scenarios passed, 1 failed, 0 skipped
24 steps passed, 1 failed, 0 skipped, 0 undefined
Took 3m56.385s

Exited with code exit status 1

CircleCI received exit code 1

This is the error:

DatasetError: An exception occurred when parsing config for dataset 
'example_classifier':
Dataset type 'kedro.io.memory_dataset.MemoryDataSet' is invalid: all data set 
types must extend 'AbstractDataSet'.

[astrojuanlu]

Root cause:

kedro-starters/pyspark-iris/{{ cookiecutter.repo_name }}/conf/base/catalog.yml

Lines 97 to 103 in 2d86c0f

	# This is an example how to use `MemoryDataSet` with Spark objects that aren't `DataFrame`'s.
	# In particular, the `assign` copy mode ensures that the `MemoryDataSet` will be assigned
	# the Spark object itself, not a deepcopy version of it, since deepcopy doesn't work with
	# Spark object generally.
	example_classifier:
	type: MemoryDataSet
	copy_mode: assign

and

https://github.com/kedro-org/kedro/blob/fd8162d5bf384ef666c01ef2c529d01fd9fa8354/kedro/io/core.py#L427-L431

[astrojuanlu]

In progress at kedro-org/kedro#2724

[astrojuanlu]

Test: gh-138