incorrect batch count for rolling update

[huangtianhua]

When try rolling update for autoscaling group, the capacity should be effective_capacity while validating whether the updation will resul in stack update timeout .

[OpenStack Infra (hudson-openstack)]

Fix proposed to branch: master
Review: https://review.openstack.org/132516

[OpenStack Infra (hudson-openstack)]

Reviewed: https://review.openstack.org/132516
Committed: https://git.openstack.org/cgit/openstack/heat/commit/?id=2edded25b216e0c160b6c2b60e1c95d0ddc29b08
Submitter: Jenkins
Branch: master

commit 2edded25b216e0c160b6c2b60e1c95d0ddc29b08
Author: huangtianhua <email address hidden>
Date: Mon Nov 3 15:27:44 2014 +0800

    Correct validation for timeout when rolling update

    Using effective capacity not the current capacity to calculate
    whether the update action will in result timeout.

    Change-Id: I718e19c197efa5ca3add270a7470ed26e0f705db
    Closes-Bug: #1388695