Add replicaset and job metaGenerator based on watcher

[ChrsMark]

Part of #31 to solve increased memory usage of add_resource_metadata.deployment and add_resource_metadata.cronjob based on what was described at #31.

The library "interface" that changes is the

elastic-agent-autodiscover/kubernetes/metadata/metadata.go

Lines 91 to 96 in f4ec42c

	func GetPodMetaGen(
	cfg *config.C,
	podWatcher kubernetes.Watcher,
	nodeWatcher kubernetes.Watcher,
	namespaceWatcher kubernetes.Watcher,
	metaConf *AddResourceMetadataConfig) MetaGen {

with https://github.com/elastic/elastic-agent-autodiscover/pull/44/files#diff-b153a7b8a2ce2b522412808eec2fa56bfa612345ca05dba323a7ef5ad5d3511cR91-R98. With this change the GetPodMetaGen requires the 2 additional watchers.

We need to adopt accordingly Beats and Agent parts so as to use the new version of the library properly:

• Beats PR: Use replicaset and Job MetaGen based on watchers beats#35483
• Agent PR: TBA

Note: I'm planning to add new tests for the 2 additional MetaGenerators in this PR however review can already start in the main codebase's changes.

Screenshots

The following screenshots indicate that the memory leak does not occur with this implementation as well as that the fields are populated properly.

[elasticmachine]

💚 Build Succeeded

the below badges are clickable and redirect to their specific view in the CI or DOCS

Expand to view the summary

Build stats

• Start Time: 2023-05-18T11:04:58.948+0000

• Duration: 11 min 21 sec

Test stats 🧪

Test	Results
Failed	0
Passed	252
Skipped	0
Total	252

🤖 GitHub comments

Expand to view the GitHub comments

To re-run your PR in the CI, just comment with:

• /test : Re-trigger the build.

[ChrsMark]

@MichaelKatsoulis @gizas this one is ready for review when you have the time :)

[MichaelKatsoulis]

Very nice work!

[gizas]

Wondering if we can put a debug message here on every put action.
And then for cases that we want to troubleshoot OOM issues we can enable debug and check what put actions might take place.

@ChrsMark would it be valuable?

[ChrsMark]

That's a valid point on high level, however I don't think a debug message at this place would actually help us. Here we just retrieve an already stored value and also this message would occur many times for big number of Pods so this would be an overkill. Sth helpful would be to check the size of the local caches (Informers' cache) but this is quite out of scope I think on this PR. In any case have created elastic/beats#35591 to evaluate further improvements.