fix(mads): add mutex when checking if reconcile is needed and reconciling

[lobkovilya]

This is a straightforward fix for the MADS problem we see in unit tests. But we probably don't need NeedsReconciliation at all since we fixed watchdogs to reconcile before the first tick #11105

CC @Automaat @michaelbeaumont

Checklist prior to review

• Link to relevant issue as well as docs and UI issues --
• This will not break child repos: it doesn't hardcode values (.e.g "kumahq" as a image registry) and it will work on Windows, system specific functions like syscall.Mkfifo have equivalent implementation on the other OS --
• Tests (Unit test, E2E tests, manual test on universal and k8s) --
  • Don't forget ci/ labels to run additional/fewer tests
• Do you need to update UPGRADE.md? --
• Does it need to be backported according to the backporting policy? (this GH action will add "backport" label based on these file globs, if you want to prevent it from adding the "backport" label use no-backport-autolabel label) --

[Automaat]

But we probably don't need NeedsReconciliation at all

I am not sure about this, MADS works in a weird way since it is responding to http requests not relying entirely on watchdogs.

[lahabana]

This is simple fix. The more complex fix is trickier because of these reconcilers running concurrently makes 0 sense. So this will work for now

[Icarus9913]

I am not sure about this, MADS works in a weird way since it is responding to http requests not relying entirely on watchdogs.

In my opinion the watchdog runs in goroutine and we need every interval to generate the XDS. And each http request gets response from go-control-plane fetch method.

Although we got that first tick reconcile from #11105, but you can imagine that the request happens between the first and second tick to fetch xds configurations then it would not found something new. That's why there's a restCallback function in MADS.