[shell] make shell ping command asynchronous

[gjc13]

Problem

Fixes #7519 shell app ping cannot receive

Change overview

This is caused by the sleep in the ping client blocking the stack's event loop.
This PR moves the timeout handling logic to the timer handlers to resolve the issue.

Testing

Manually tested with chip-shell and chip-echo-responder

[andy31415]

should we at some point add interface id arguments?
If we start to use link-local addreses, the interface will matter.

likely not for this PR.

[yufengwangca]

The conversation has been closed, and I am not able to reopen it, suppose the first SendEchoRequest succeed, and the second SendEchoRequest failed with error within EchoTimerHandler. How Shutdown() is triggered in this case?

[gjc13]

The conversation has been closed, and I am not able to reopen it, suppose the first SendEchoRequest succeed, and the second SendEchoRequest failed with error within EchoTimerHandler. How Shutdown() is triggered in this case?

Each SendEchoRequet will schedule the next send in the timer handler by calling StartTimer. If the SendEchoRequest fails it will cancel the next send by calling CancelTimer. The ping will be aborted once any of the send fails just as the behavior in the original version.

[woody-apple]

/rebase

[github-actions]

Size increase report for "nrfconnect-example-build" from 1397b4b

File	Section	File	VM
chip-shell.elf	bss	0	1280
chip-shell.elf	text	128	128

Full report output

BLOAT REPORT

Files found only in the build output:
    report.csv

Comparing ./master_artifact/chip-shell.elf and ./pull_artifact/chip-shell.elf:

sections,vmsize,filesize
bss,1280,0
.debug_info,0,281
.strtab,0,202
.symtab,0,192
.debug_loc,0,129
.debug_ranges,0,128
text,128,128
.debug_line,0,121
.debug_frame,0,108
.debug_abbrev,0,70
.debug_aranges,0,16
.debug_str,0,11
.shstrtab,0,-2

Comparing ./master_artifact/chip-lock.elf and ./pull_artifact/chip-lock.elf:

sections,vmsize,filesize