Try detecting timer and activity resurrection

[yycptt]

What changed?
When processing user timer or activity timeout task, if a timer is overdue for a long time (by default, 24 hours), scan workflow history to see if that timer has already been fired or if that activity has ready finished.

Why?
For detecting timer/activity resurrection and prevent duplicated timer fired or activity completion.

How did you test it?

• Unit test.

Potential risks
N/A in the worst case, we can increase the threshold for detecting resurrection to be a large number to turn the feature off.

Release notes

Documentation Changes

[coveralls]

Pull Request Test Coverage Report for Build ce964594-29fe-49e4-a9df-52f9658ba3c5

• 162 of 190 (85.26%) changed or added relevant lines in 4 files are covered.
• 154 unchanged lines in 15 files lost coverage.
• Overall coverage decreased (-0.02%) to 56.478%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
service/history/task/timer_active_task_executor.go	150	178	84.27%

Files with Coverage Reduction	New Missed Lines	%
common/membership/rpServiceResolver.go	2	73.58%
common/task/weightedRoundRobinTaskScheduler.go	2	88.6%
service/history/execution/mutable_state_builder.go	2	69.86%
service/matching/taskListManager.go	2	74.09%
common/persistence/nosql/nosqlplugin/cassandra/workflow.go	3	50.23%
service/history/queue/timer_gate.go	3	95.83%
common/persistence/executionManager.go	4	75.5%
common/persistence/statsComputer.go	4	95.71%
common/persistence/nosql/nosqlplugin/cassandra/workflowUtils.go	11	77.83%
service/history/queue/transfer_queue_processor.go	12	57.64%

Totals
Change from base Build 4a34a039-15ff-46ba-b5f7-0f167c6f8e52:	-0.02%
Covered Lines:	78896
Relevant Lines:	139693

---

💛 - Coveralls

[yux0]

Great changes. I think we can also have a similar thing to detect the mutable state and history not sync issue.