fix(sdk): always re-evaluate termination after checkpoint queue drains

[ayushiahjolia]

Issue #, if available: #510

Description of changes:
Removes the if (!this.terminationTimer) guard in processQueue's finally block so that termination is always re-evaluated after the checkpoint queue drains.

Problem

In map + waitForCallback scenarios, callbacks register as IDLE_AWAITED and set terminationTimer before a completed branch's fire-and-forget checkpoint enters the queue. When the queue drains, the guard sees terminationTimer is
non-null and skips checkAndTerminate. The function never suspends - it wedges until an unrelated timer (e.g., a sibling ctx.wait) fires.

Fix

Replace the conditional with an unconditional this.checkAndTerminate().
This is safe because scheduleTermination already returns early when the same reason is already scheduled — no timer restart, no extra cooldown in the common case.

Testing

• Added unit tests that fail without the fix and pass with it
• Added map + waitForCallback and map + waitForCondition example tests
• All existing tests pass

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

[ParidelPooya]

I ran new added examples and integration test with and without the fix (change to checkpoint-manager), ignoring assertEventSignatures, integration test are passing before and after the change.

That means that we didn't manage to reproduce the customer issue.