Integrate KVS with device session tables

[pan-apple]

Problem

The previous PR (#4952) was reverted due to a crash loop on the device.

Summary of Changes

This PR fixes the crash issue caused by #4952. The crash was due to using MemoryAlloc instead of New<> for allocating PASESession objects. It caused memory to be allocated, but the object was not initialized.

The old PR added the following functionality:
Added code to store relevant information for Admin and Peer Connection/PASE Session in KVS storage. The device will read KVS and recreate the tables at boot up. Tested the change on m5stack.

[todo]

The admin ID space allocation should be re-evaluated. With the current approach, the space could be

connectedhomeip/src/app/server/Server.cpp

Lines 130 to 140 in 4ada40b

	// TODO: The admin ID space allocation should be re-evaluated. With the current approach, the space could be
	// exhausted while IDs are still available (e.g. if the admin IDs are allocated and freed over a period of time).
	// Also, the current approach can make ID lookup slower as more IDs are allocated and freed.
	for (AdminId id = 0; id < nextAvailableId; id++)
	{
	AdminPairingInfo * admin = adminPairings.AssignAdminId(id);
	// Recreate the binding if one exists in persistent storage. Else skip to the next ID
	if (admin->FetchFromKVS(gServerStorage) != CHIP_NO_ERROR)
	{
	adminPairings.ReleaseAdminId(id);
	}

---

This comment was generated by todo based on a TODO comment in 4ada40b in #4976. cc @pan-apple.

[pan-apple]

rebased

[github-actions]

Size increase report for "nrfconnect-example-build" from 7c09658

File	Section	File	VM
chip-lighting.elf	text	2016	2016
chip-lighting.elf	rodata	552	556
chip-lighting.elf	[LOAD #3 [RW]]	0	31
chip-lighting.elf	bss	0	1
chip-lock.elf	text	2016	2016
chip-lock.elf	rodata	560	556

Full report output

BLOAT REPORT

Files found only in the build output:
    report.csv

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

sections,vmsize,filesize
.debug_info,0,101149
.debug_line,0,12555
.debug_abbrev,0,9628
.debug_loc,0,8202
.debug_str,0,5538
.strtab,0,2789
text,2016,2016
.debug_ranges,0,1728
.symtab,0,1712
.debug_frame,0,928
rodata,556,552
.debug_aranges,0,336
[LOAD #3 [RW]],31,0
.shstrtab,0,3
bss,1,0

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

sections,vmsize,filesize
.debug_info,0,101149
.debug_line,0,12559
.debug_abbrev,0,9628
.debug_loc,0,8198
.debug_str,0,5538
.strtab,0,2789
text,2016,2016
.debug_ranges,0,1728
.symtab,0,1712
.debug_frame,0,928
rodata,556,560
.debug_aranges,0,336
.shstrtab,0,3

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

sections,vmsize,filesize

[github-actions]

Size increase report for "gn_qpg6100-example-build" from 7c09658

File	Section	File	VM
chip-qpg6100-lock-example.out	.text	1352	1352
chip-qpg6100-lock-example.out	.bss	0	16
chip-qpg6100-lock-example.out	.data	-4	-4
chip-qpg6100-lock-example.out	.heap	0	-16

Full report output

BLOAT REPORT

Files found only in the build output:
    report.csv

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

sections,vmsize,filesize
.debug_info,0,50366
.debug_line,0,8491
.debug_abbrev,0,6118
.debug_str,0,4369
.debug_loc,0,3104
.strtab,0,2136
.symtab,0,1440
.text,1352,1352
.debug_frame,0,908
.debug_aranges,0,320
.debug_ranges,0,176
[ELF Headers],0,32
.bss,16,0
.data,-4,-4
.heap,-16,0
[Unmapped],0,-1384

[rgoliver]

Maybe this comment needs to be updated to state that every function returns NOT_IMPLEMENTED (same on other platforms), or maybe just a comment/TODO on the class to explain.