Fix cets_status improper missing/unknown tables

[arcusfelis]

This PR addresses MIM-2092.

systemInfo looks like that when testing in k8s with Helm:

kubectl exec -it mongooseim-0 -- mongooseimctl cets systemInfo
{
  "data" : {
    "cets" : {
      "systemInfo" : {
        "unavailableNodes" : [

        ],
        "remoteUnknownTables" : [
          "cets_s2s_secret",
          "cets_s2s_session",
          "cets_s2s_secret",
          "cets_s2s_session",
          "cets_s2s_secret",
          "cets_s2s_session",
          "cets_s2s_secret",
          "cets_s2s_session",
          "cets_s2s_secret",
          "cets_s2s_session",
          "cets_s2s_secret",
          "cets_bosh",
          "cets_session",
          "cets_bosh",
          ...],
                  "remoteNodesWithUnknownTables" : [
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]"
        ],
        "remoteNodesWithMissingTables" : [
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]",
          "[email protected]"
        ],
        "remoteMissingTables" : [
          "cets_bosh",
          "cets_cluster_id",
          "cets_s2s_secret"
        ],
        ...

Why? Because maps:keys/1 returns sorted keys only in tests (because atoms used in modules are sorted at the compilation step, so it is hard to find this bug in the synthetic environment. Test would also "pass" if we just try to parse status_data.txt and pass it into cets_status:format_data/1, without first tweaking the atom table like in cets_SUITE:test_data_for_duplicate_missing_table_in_status/1 function).
Once you start registering atoms dynamically, maps:keys/1 returns keys in the order atoms were registered in the atom table.

This PR ensures that keys are sorted before we do following calculations (like determining which tables are missing).

This PR also splits cets_status into gather_data and format_data, so we can more easily debug such issues.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Comparison is base (5fb3db1) 97.98% compared to head (e28f49e) 98.00%.

Additional details and impacted files

@@            Coverage Diff             @@
##             main      #40      +/-   ##
==========================================
+ Coverage   97.98%   98.00%   +0.02%     
==========================================
  Files           9        9              
  Lines         645      652       +7     
==========================================
+ Hits          632      639       +7     
  Misses         13       13              

Files	Coverage Δ
src/cets_status.erl	100.00% <100.00%> (ø)

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[pawlooss1]

Looking good 😎 This was surely a tricky one. I also find split version of status/1 easier to read.

[chrzaszcz]

I'm a bit late, but it looks good to me as well 🙂