DRIVERS-2951 Test serverMonitoringMode=poll waits after a successful heartbeat

[ShaneHarvey]

Fixes DRIVERS-2951.

Please complete the following before merging:

• Make sure there are generated JSON files from the YAML test files.
• Test changes in at least one language driver.
• Test these changes against all server versions and topologies (including standalone, replica set, sharded
  clusters, and serverless).

Question, should we still update the prose test here added in DRIVERS-1251 to run on both streaming and polling? Currently it implicitly tests polling on <4.4 servers.

[ShaneHarvey]

@prestonvasquez could you review this to ensure it covers the Go driver bug? When I introduce the bug in Python the test fails like this:

TestUnifiedServerMonitoringMode.test_poll_waits_after_successful_heartbeat
...
test/unified_format.py:1635: in _testOperation_assertEventCount
    self.assertEqual(self._event_count(client, event), count, f"expected {count} not {event!r}")
E   AssertionError: 5169 != 1 : expected 1 not {'serverHeartbeatStartedEvent': {}}

[prestonvasquez]

@prestonvasquez could you review this to ensure it covers the Go driver bug? When I introduce the bug in Python the test fails like this:

TestUnifiedServerMonitoringMode.test_poll_waits_after_successful_heartbeat
...
test/unified_format.py:1635: in _testOperation_assertEventCount
    self.assertEqual(self._event_count(client, event), count, f"expected {count} not {event!r}")
E   AssertionError: 5169 != 1 : expected 1 not {'serverHeartbeatStartedEvent': {}}

@ShaneHarvey Go has not implemented DRIVERS-2366 and so we don't have wait or assertEventCount. Though tests described in this PR do appear to work for POC mongodb/mongo-go-driver#1731 and fail when ran against the bug:

=== RUN   TestUnifiedSpec/server-discovery-and-monitoring/unified/serverMonitoringMode.json/poll_waits_after_successful_heartbeat
    unified_spec_runner.go:169: error running operation "assertEventCount" at index 3: execution failed: failed to assert count. want: 1, got: 8642
--- FAIL: TestUnifiedSpec (0.60s)
    --- FAIL: TestUnifiedSpec/server-discovery-and-monitoring/unified (0.60s)
        --- FAIL: TestUnifiedSpec/server-discovery-and-monitoring/unified/serverMonitoringMode.json (0.60s)
            --- FAIL: TestUnifiedSpec/server-discovery-and-monitoring/unified/serverMonitoringMode.json/poll_waits_after_successful_heartbeat (0.60s)
FAIL
exit status 1
FAIL    go.mongodb.org/mongo-driver/mongo/integration/unified   1.856s

[W-A-James]

Seems like the file runs this on standalones, replicasets and sharded clusters. It seems that we'd either want to write separate tests for replsets and sharded clusters or we want to restrict this to run against standalones.

[ShaneHarvey]

We already have this at the top of the file:

# These tests cannot run on replica sets because the order of the expected
# SDAM events are non-deterministic when monitoring multiple servers.
# They also cannot run on Serverless or load balanced clusters where SDAM is disabled.
runOnRequirements:
  - topologies: [single, sharded, sharded-replicaset]
    serverless: forbid

This test works on both standalone and sharded clusters.