IPPool doesn't free IP-addresses if healing failed

[glazychev-art]

Current Behavior

This behavior is related to healing.
For example, if local icmp-responder restarted - we start healing from nsmgr.
New icmp-responder receives new request and pulls IP addresses. But if requestTimeout is not enough for icmp-responder, nsmgr will close context with context deadline exceeded error before the request comes back. In this case we will lose allocated IP-addresses

Expected Behavior

Free allocated IP-addresses

Failure Information (for bugs)

For the next request we get:

...
2021-07-02T12:48:48.848275423Z Jul  2 12:48:48.848�[31m [ERRO] [name:nse-kernel-6d67d65dc6-vstxz] �[0m(9.1)           IPPool is empty
...

Steps to Reproduce

1. Step 1
2. Step 2
3. You get the idea...

[denis-tingaikin]

Seems like duplicate of #1020

[glazychev-art]

This refers to the previous implementation of healing.
I think this issue can be closed