Add closer server to the NSMgr chain

[Bolodya1997]

Issue

Client performs Close with some context timeout. It can be enough to reach NSMgr but in some cases (e.g. on high load) it can be not enough to pass the whole networkservice chain across all hops, so this Close would fail even it has reach the NSMgr.

Solution

Add closer server in the beginning of the NSMgr networkservice chain, so it would restart Close event on NSMgr application context with the bigger timeout.

References

#1031

[edwarnicke]

@Bolodya1997 Isn't this already taken care of by the Refresh timer?

[Bolodya1997]

I guess this should be resolved by networkservicemesh/deployments-k8s#906, so we can close this issue.