Running salt-cloud within orchestration always reports success

Description of Issue/Question

When running cloud functions within orchestrate, orchestrate can't catch failures correctly.

The issue is that these lines in salt/states/saltmod.py work most of the time:
if isinstance(runner_return, dict) and 'Error' in runner_return:
out['success'] = False

These stop working when the return dict element is itself a dictionary, like this line in my example:

[INFO ] {u'return': {u'servername': {u'Error': u'servername already exists under mh-vmware:vmware'}}}

Setup

#Spins up a new host taking the last octet of the IP as an argument
'provision_single_host':
salt.runner:
- name: cloud.profile
- prof: single_server
- instances:
- servername{{pillar.ip}}
- vm_overrides:
devices:
network:
Network adapter 1:
name: VL1011_TestNet1
switch_type: standard
adapter_type: vmxnet3
ip: ip{{pillar.ip}}
gateway: [gateway]
subnet_mask: 255.255.255.0

Steps to Reproduce Issue

The easiest way to reproduce this is to have an orchestration that creates a new server when that server already exists

sudo salt-run state.orchestrate orchestrate.test_single_server pillar='{"ip": "200"}' saltenv=*testenv* -l info
[INFO    ] Loading fresh modules for state activity
[INFO    ] Fetching file from saltenv '*testenv*', ** done ** 'orchestrate/test_single_server.sls'
[INFO    ] Running state [cloud.profile] at time 11:43:51.146055
[INFO    ] Executing state salt.runner for [cloud.profile]
[ERROR   ] *servername* already exists under mh-vmware:vmware
[INFO    ] Runner completed: 20181220114351486434
[INFO    ] {u'return': {u'*servername*': {u'Error': u'*servername* already exists under mh-vmware:vmware'}}}
[INFO    ] Completed state [cloud.profile] at time 11:43:52.064214 (duration_in_ms=918.16)
*saltmaster*:
----------
          ID: provision_single_host
    Function: salt.runner
        Name: cloud.profile
      Result: True
     Comment: Runner function 'cloud.profile' executed.
     Started: 11:43:51.146054
    Duration: 918.16 ms
     Changes:
              ----------
              return:
                  ----------
                  *servername*:
                      ----------
                      Error:
                          *servername* already exists under mh-vmware:vmware

Summary for *saltmaster*
------------
Succeeded: 1 (changed=1)
Failed:    0
------------
Total states run:     1
Total run time: 918.160 ms
[INFO    ] Runner completed: 20181220114347984639

Versions Report

salt --versions-report
Salt Version:
Salt: 2018.3.0

Dependency Versions:
cffi: 1.11.2
cherrypy: unknown
dateutil: Not Installed
docker-py: Not Installed
gitdb: 0.6.4
gitpython: Not Installed
ioflo: Not Installed
Jinja2: 2.7.2
libgit2: 0.26.3
libnacl: Not Installed
M2Crypto: 0.28.2
Mako: Not Installed
msgpack-pure: Not Installed
msgpack-python: 0.5.1
mysql-python: Not Installed
pycparser: 2.14
pycrypto: 2.6.1
pycryptodome: Not Installed
pygit2: 0.26.4
Python: 2.7.5 (default, Apr 11 2018, 07:36:10)
python-gnupg: Not Installed
PyYAML: 3.11
PyZMQ: 15.3.0
RAET: Not Installed
smmap: 0.9.0
timelib: Not Installed
Tornado: 4.2.1
ZMQ: 4.1.4

System Versions:
dist: centos 7.5.1804 Core
locale: UTF-8
machine: x86_64
release: 3.10.0-862.3.2.el7.x86_64
system: Linux
version: CentOS Linux 7.5.1804 Core

Start in salt/states/saltmod.py, focusing on how the salt.runner state evaluates the runner return, then run the provided salt-run state.orchestrate reproduction. Done means an Error nested inside the cloud.profile return is recognized as a failure instead of reporting Result: True.