fix(Core): Fixing concurrency issues in AWSSynchronizedMutableDictionary

[sebaland]

Issue #, if available:

• [AWSSynchronizedMutableDictionary removeObjectForKey:] crashed in ver 2.36.6 #5412

Description of changes:

When using barriers and concurrent queues, doing dispatch_barrier_sync can cause deadlocks (and crashes), so dispatch_barrier_async is recommended.
Our AWSSynchronizedMutableDictionary was using sync for write operations, which are really unnecessary and can be async. Read operations do remain sync, but with no barrier involved.

---

Also included in this PR is a potential fix for the random crash when disconnecting and quickly connecting again. If a disconnection happens due to an error, the reconnectThread is trigger, which sets the reconnectTimer so that a reconnection is attempted in x seconds.
But if a new connection is explicitly triggered, a new streamsThread will be created and the old one discarded. We do attempt to invalidate the reconnectTimer in these situations, but we try to do it in the reconnectThread as that is where the timer was set.

However, if the reconnectThread has finished (i.e. not running), then the timer will not be invalidated and will fire eventually, which will attempt to use the now deallocated old streamsThread.
So I'm just adding a simple check to allow us to invalidate the timer anyway if the reconnectThread has finished.

---

Check points:

• Added new tests to cover change, if needed
• All unit tests pass
• All integration tests pass
• Updated CHANGELOG.md
• Documentation update for the change if required
• PR title conforms to conventional commit style

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.