Add integ test for Amazon Managed Prometheus as destination

[movence]

Depends on aws/amazon-cloudwatch-agent#1192

Description of changes

Add integ test for Amazon Managed Prometheus as destination.

New AMP integ test:

• Uses existing linux terraform to set up an ec2 stack which also creates AMP workspace when test/amp is used
• Uses CPU and Process plugins to emit and validate metrics
• validates metrics with append_dimensions (both metric and global level), rename and aggregation_dimensions

License

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

Tests

Test run:

null_resource.integration_test_run (remote-exec): This is the aws region:  us-west-2
null_resource.integration_test_run (remote-exec): === RUN   TestAmp
null_resource.integration_test_run (remote-exec): 2024/07/25 20:54:46 Running AMPTest
...
null_resource.integration_test_run (remote-exec): 2024/07/25 20:55:50 The average value 0.333333 for metric processes_zombies are within bound [0.000000, 0.000000]
null_resource.integration_test_run (remote-exec): --- PASS: TestAmp (64.29s)
null_resource.integration_test_run (remote-exec): PASS
null_resource.integration_test_run (remote-exec): ok  	github.com/aws/amazon-cloudwatch-agent-test/test/amp	64.314s
null_resource.integration_test_run: Creation complete after 10m6s [id=6863983178346704537]

[Paramadon]

Why do we need to drop these two metrics?

[Paramadon]

LGTM!

[okankoAMZ]

My only comment is can we make the test name bit more descriptive than AmpTest . It feels like it will be hard to distinguish in the long run

[okankoAMZ]

LGTM!

[movence]

Using PromQL function avg_over_time for last 3m https://prometheus.io/docs/prometheus/latest/querying/functions/#aggregation_over_time