Description of problem:
 - ceph-volume fails to zap the lvm device(advanced configuration)
 - While executing "ceph-volume lvm zap --destroy" against lvm device, executing dd on the lvm device. But after that, it tries to remove vg and lv. 
---
[ceph: root@m2 /]# ceph-volume lvm zap --destroy /dev/vg_mpatha/lv_mpatha
--> Zapping: /dev/vg_mpatha/lv_mpatha
Running command: /usr/bin/dd if=/dev/zero of=/dev/vg_mpatha/lv_mpatha bs=1M count=10 conv=fsync
 stderr: 10+0 records in
10+0 records out
 stderr: 10485760 bytes (10 MB, 10 MiB) copied, 0.0631291 s, 166 MB/s
--> Only 1 LV left in VG, will proceed to destroy volume group vg_mpatha
Running command: /usr/sbin/vgremove -v -f vg_mpatha
 stderr: Removing vg_mpatha-lv_mpatha (253:3)
 stderr: Archiving volume group "vg_mpatha" metadata (seqno 3).
  Releasing logical volume "lv_mpatha"
 stdout: Logical volume "lv_mpatha" successfully removed
 stderr: Creating volume group backup "/etc/lvm/backup/vg_mpatha" (seqno 4).
 stderr: Removing physical volume "/dev/mapper/mpatha" from volume group "vg_mpatha"
 stdout: Volume group "vg_mpatha" successfully removed
--> Zapping successful for: <LV: /dev/vg_mpatha/lv_mpatha>
---

Version-Release number of selected component (if applicable):
 RHCS 5.x

How/steps to reproduce:
 - Create lvm and deploy the osd using "ceph orch daemon add osd hostname:vg/lv"
 - Remove the osd.
 - zap the lvm device

Actual results:
 - dd on lvm device
 - lv is removing. 
 - vg is removing.
 - Disk header and lvm label remains persists.
 

Expected results:
 - This should not remove the vg. (the vg might contain other lv since this is advanced lvm scenario)
 - ceph-volume should handle the command "ceph-volume lvm zap --destroy" properly.