Fix 8.0.0 related issues in E2E

[david-kow]

This PR fixes a number of issues that prevent E2E tests from passing on 8.0.0-alpha2:

• ignore all suffixes (not only -SNAPSHOT) when checking APM Server version
• don't provide ent_search.auth.* settings in Enterprise Search config for 8.0.0+
• use monitoring.ui.container.elasticsearch.enabled instead of xpack.monitoring.ui.container.elasticsearch.enabled in Kibana config for 7.16.0+
• use node.roles instead of node.data: true etc. in Elasticsearch config for7.9.0+ E2E tests and manifests
• set kibana.host setting in Enterprise Search config for 8.0.0+ E2E tests
• inspect metricbeat-* instead of .monitoring-kibana-* index for checking Stack Monitoring data in Elasticsearch and Kibana
• check for level == available instead of state == green field when checking Kibana health
• check /internal/workplace_search/overview instead of /api/enterprise_search/config_data API when checking Enterprise Search UI availability in Kibana
• Temporarily disable TestVolumeMultiDataPath test for 8.x as the feature deprecation revert is still in progress

[david-kow]

jenkins test this please

[david-kow]

run full pr build

[david-kow]

Tests look good, the only failure (below) is caused by using nodes.roles notion in pre-7.9.0 ES versions.

This is fixed in 455e60c.

[david-kow]

jenkins test this please

07:48:47          	            				handler_integration_test.go:82
07:48:47          	Error:      	Received unexpected error:
07:48:47          	            	failed waiting for all runnables to end within grace period of 30s: context deadline exceeded
07:48:47          	Test:       	TestDynamicEnqueueRequest```

[pebrc]

LGTM