Don't automatically close exchanges if the underlying connection closes.

[bzbarsky-apple]

Someone might still have a reference to those exchanges, and we will
end up with an unbalanced release. Instead just notify our consumer
in the specific case when we are waiting for a response so it knows
the response will never come (and presumably closes the exchange as
needed).

Problem

Fixes #7012

Change overview

Instead of hard-closing exchanges on connection close, just notify the ones that are waiting for a response that it won't be coming.

Testing

• There are automated tests for some basic behavior included in this PR.
• I manually tested that applying this on top of [controller] Use short lived CommandSender when no command sender is provided #7041 allows me to successfully complete ble-wifi pairing with the C++ chip-tool.

[github-actions]

Size increase report for "esp32-example-build" from 06d606f

File	Section	File	VM
chip-all-clusters-app.elf	.flash.text	92	92

Full report output

BLOAT REPORT

Files found only in the build output:
    report.csv

Comparing ./master_artifact/chip-all-clusters-app.elf and ./pull_artifact/chip-all-clusters-app.elf:

sections,vmsize,filesize
.debug_info,0,3501
.debug_line,0,256
.debug_str,0,142
.strtab,0,120
.flash.text,92,92
.debug_loc,0,63
.debug_frame,0,48
.debug_abbrev,0,32
.symtab,0,32
.debug_aranges,0,16
.debug_ranges,0,16
.xt.prop._ZTVN4chip11DeviceLayer37DeviceNetworkProvisioningDelegateImplE,0,2

Comparing ./master_artifact/chip-pigweed-app.elf and ./pull_artifact/chip-pigweed-app.elf:

sections,vmsize,filesize

[github-actions]

Size increase report for "gn_qpg6100-example-build" from 06d606f

File	Section	File	VM
chip-qpg6100-lighting-example.out	.text	96	96

Full report output

BLOAT REPORT

Files found only in the build output:
    report.csv

Comparing ./master_artifact/chip-qpg6100-lighting-example.out.map and ./pull_artifact/chip-qpg6100-lighting-example.out.map:

BLOAT EXECUTION FAILED WITH CODE 1:
bloaty: unknown file type for file './pull_artifact/chip-qpg6100-lighting-example.out.map'

Comparing ./master_artifact/chip-qpg6100-lighting-example.out and ./pull_artifact/chip-qpg6100-lighting-example.out:

sections,vmsize,filesize
.debug_info,0,2762
.debug_loc,0,456
.debug_line,0,181
.debug_str,0,141
.strtab,0,120
.text,96,96
.symtab,0,64
.debug_frame,0,48
.debug_ranges,0,40
.debug_aranges,0,16
[Unmapped],0,-96

[github-actions]

Size increase report for "nrfconnect-example-build" from 06d606f

File	Section	File	VM
chip-lock.elf	text	88	88
chip-lock.elf	device_handles	8	8
chip-shell.elf	text	88	88
chip-shell.elf	device_handles	-8	-8

Full report output

BLOAT REPORT

Files found only in the build output:
    report.csv

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

sections,vmsize,filesize
.debug_info,0,2814
.debug_loc,0,382
.debug_line,0,182
.debug_str,0,142
.strtab,0,120
text,88,88
.symtab,0,64
.debug_frame,0,48
.debug_ranges,0,40
.debug_aranges,0,16
device_handles,8,8

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

sections,vmsize,filesize
.debug_info,0,942
.debug_loc,0,458
.debug_line,0,178
.debug_str,0,142
.strtab,0,120
text,88,88
.symtab,0,64
.debug_frame,0,48
.debug_ranges,0,40
.debug_aranges,0,16
device_handles,-8,-8

[yufengwangca]

After we reset the SessionHandle, this exchange can not be used to send any message.

Let's go back the the original issue.

After we send AddRootCert command and receive the response to AddRootCert, the connection is closed.
Now we don't close the exchange, but reset the SessionHandle of the exchange.

On which exchange we send AddNetwork Command? My understand is we should continue to use the same exchange to send this command and close the exchange when we reach the end of conversation.

[bzbarsky-apple]

On which exchange we send AddNetwork Command?

A different one. Each new Invoke is a new exchange in the spec.