sync: smoother manager scoping (fixes #9031)

[dogi]

Refactors SyncManager to accept an @ApplicationScope CoroutineScope via its constructor, replacing the locally created syncScope.

This change centralizes coroutine management, improves testability, and aligns with Hilt's dependency injection best practices. The ServiceModule has been updated to provide the application-level scope to SyncManager.

---

https://jules.google.com/session/3228009531546273882

[qltysh]

❌ 1 blocking issue (1 total)

Tool	Category	Rule	Count
qlty	Structure	Function with many parameters (count = 6): provideSyncManager	1	❌

[chatgpt-codex-connector]

💡 Codex Review

myplanet/app/src/main/java/org/ole/planet/myplanet/service/SyncManager.kt

Lines 112 to 115 in 487b520

	private fun destroy() {
	if (betaSync) {
	syncScope.cancel()
	ThreadSafeRealmHelper.closeThreadRealm()

Stop cancelling shared application scope

After this refactor SyncManager receives the global @ApplicationScope CoroutineScope, but destroy() still calls syncScope.cancel() when betaSync is true. Because AuthSessionUpdater and NetworkUtils inject the same @ApplicationScope scope, the first beta sync teardown would cancel the shared scope and permanently stop their background coroutines (network monitoring, auth session refreshes, etc.) for the lifetime of the process. Consider cancelling a child scope owned by SyncManager instead of the application-wide scope.

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".