Don't lose VM volume attachments when refreshing the cloud inventory

[mansam]

When the CloudManager refreshes, it removes from the inventory any Disks that were only added by the CinderManager as part of attaching Volumes to VMs since the CloudManager doesn't find them while collecting inventory. As a result, Volumes will have correct statuses but incorrect numbers of attachments, leading to confusing user experience and various UI actions failing.
This patch causes the CloudManager to be aware of VM volume attachments so that the associations are not broken between Cloud and Cinder refreshes.

This is a solution for https://bugzilla.redhat.com/show_bug.cgi?id=1544344

Depends on #242

[miq-bot]

Checked commits mansam/manageiq-providers-openstack@d079d62~...703d1f0 with ruby 2.3.3, rubocop 0.52.1, haml-lint 0.20.0, and yamllint 1.10.0
12 files checked, 4 offenses detected

**

• 💣 💥 🔥 🚒 - Linter/Yaml - missing config files

app/models/manageiq/providers/openstack/inventory/parser/cloud_manager.rb

• ❗ - Line 391, Col 10 - Style/NumericPredicate - Use s.attributes.fetch("os-extended-volumes:volumes_attached", []).length.positive? instead of s.attributes.fetch("os-extended-volumes:volumes_attached", []).length > 0.
• ❗ - Line 391, Col 10 - Style/ZeroLengthPredicate - Use !empty? instead of length > 0.
• ❗ - Line 391, Col 7 - Style/Next - Use next to skip iteration.

[aufi]

Looks good to me, merging!