[sheepdog] [ANNOUNCE] sheepdog stable release v0.9.1_rc0

[Hitoshi Mitake]

Hi sheepdog users and developers,

I released v0.9.1_rc0 of the stable branch. You can download a source
archive from here:
https://github.com/sheepdog/sheepdog/releases/tag/v0.9.1_rc0

This update contains various bugfixes. Important topics are:
- bugfixes related to iSCSI multipath
- systemd script update
- data wipe bug caused by gateway nodes

If no one disagrees about this release candidate in 2 days, it will be
v0.9.1 officially.

Below is the summary of the updates:

Bingpeng Zhu (1):
      sheep/http: check correct variable

Hitoshi Mitake (4):
      sheep: invalidate inode status of other nodes before actual write
      sheep: handle a case of small buffer in cinfo collection
      sheep: use correct number of copies during reference decrement
      sheep: let gateway node exit in a case of gateway only cluster

Jinzhi Chen (3):
      init trace for work queue before journal init
      init trace for work queue in dog command
      sheep/ops.c: fix problem caused by uninitialized backend storage in gateway

Liu Yuan (1):
      corosync: fix a compiler warning

Saeki Masaki (1):
      dog: referencing uninitialized buffer during cluster format

Teruaki Ishizaki (1):
      dog: fix segfault bug of getting snapshot for "LOCK_STATE_SHARED" VDI.

Vladislav Gorbunov (2):
      script: fix systemd support
      use systemd restart ability to monitor sheepdog daemon

Xiaolong Xu (1):
      sheep: remove the duplicate `if` statement

YAMADA Hideki (1):
      dog/vdi: enable to unlock a vdi locked by tgtd

Thanks,
Hitoshi