Test cases for virsh iface related commands

[will-Do]

@liyangfnst @yangdongsheng @yumingfei
Cc @kylazhang
Hi, We are planning to implement virsh iface related test cases. As there's an closed PR(#681) about this issue, i think it's better to consult with you guys. Any feedback would be appreciated.
Thanks.

[will-Do]

As most of these commands has no option, it's better to implement them in one logical case, just like pool related commands:
Case 1: iface-define -> iface-start -> iface-list -> iface-dumpxml -> iface-mac -> iface-name -> iface-destroy -> iface-undefine
Case 2: iface-bridge -> iface-unbridge

[yumingfei]

@will-Do, thanks for reminder~
In our perspective, we need to test more relative features.
Such as, interface type when created, it may be bridge, ethernet or some else;
how to configure its IP, static or dhcp; or trying to use created interface on vms.

So if you reserve places for extending these features, we'll be ok.

[will-Do]

@yumingfei
Well, what i am doing is make sure these virsh commands work well(base iface operation, using loopback device as default).
As this series commands manipulate host interfaces, it may affect the network of the host where the case running, so we need make sure the interface we use in test will not broke host network, and that's why i'm using 'lo' by default :)
For related feature you mentioned, i believe they all can be done by iface-define, which depended on the XML file you give.