[AWS] Image operations support in Cinder

[Sanket Sudake]

Currently we don't have support for copy image to volume and volume to image.

https://github.com/openstack/omni/blob/4aa67e3eea9596f17f8b304255dec7503a4f82b4/cinder/volume/drivers/aws/ebs.py#L239

[Pratik Shah (pratikshah)]

Creation of volume from volume was also not supported in AWS. Added support for that as well in this ticket.

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/495501
Committed: https://git.openstack.org/cgit/openstack/omni/commit/?id=59af73e47b1e1a8c169ef79e060b4ff7a32645dd
Submitter: Jenkins
Branch: master

commit 59af73e47b1e1a8c169ef79e060b4ff7a32645dd
Author: Pratik Shah <email address hidden>
Date: Sat Aug 19 15:37:30 2017 +0530

    [AWS] Added a support to create volume from image and volume from volume

    Description:
    - Changes in 'glance/glance_store/_drivers/aws.py' are required because
    when glanceclient tries to retrieve image details, it expects 2 values.
    But in the current case, it was yielding a tuple which was throwing
    exception as it was getting only 1 value when 'get()' was called

    - Added a function "clone_image()" which takes volume details, image
    metadata, object of glance service, image location as arguments.

    - Steps:
    1. get image details from AWS using image id provided in image metadata
    and get snapshot id from response. If snapshot id is None, then raise an
    exception
    2. create a dict object which contains size, zone and snapshot id as
    fields
    3. call create_volume() with dict created as argument
    4. create tags for volume created

    Closes-Bug: #1710046

    Change-Id: I241e55f45d27e14b7328ae276ec544e5360e1ebd