fix: custom crd not printing streams logs

[beastpu]

Description
my company uses a custom CRD (openkruise https://openkruise.io/) to manage workloads.
During local development, we discovered that aggregated logs were not being printed. Aggregated logs are crucial for microservices architecture.

skaffold.yaml

apiVersion: skaffold/v4beta7
kind: Config
build:
  artifacts:
  - image: skaffold-example
  local:
    push: false
manifests:
  rawYaml:
  - sts.yaml

resourceSelector:
  allow:
    - groupKind: CloneSet.apps.kruise.io
      image:
        - .*
      labels:
        - .*
    - groupKind: StatefulSet.apps.kruise.io
      image:
        - .*
      labels:
        - .*

sts.yaml

apiVersion: apps.kruise.io/v1beta1
kind: StatefulSet
metadata:
  name: getting-started
spec:
  replicas: 1
  serviceName: getting-started
  selector:
    matchLabels:
      app: getting-started
  template:
    metadata:
      labels:
        app: getting-started
    spec:
      containers:
      - name: getting-started
        image: skaffold-example
        imagePullPolicy: IfNotPresent

User facing changes (remove if N/A)
I found that the reason was the absence of corresponding GVK (GroupVersionKind) in manifest.TransformAllowlist and manifest.TransformDenylist.

My solution is to pass ConsolidateTransformConfiguration into the NewResourceSelectorImages method because we can configure resourceSelector in the skaffold.yaml file.

[codecov]

Codecov Report

Attention: 295 lines in your changes are missing coverage. Please review.

Comparison is base (290280e) 70.48% compared to head (2e15716) 63.71%.
Report is 1068 commits behind head on main.

Files	Patch %	Lines
cmd/skaffold/app/cmd/exec.go	16.32%	40 Missing and 1 partial ⚠️
cmd/skaffold/app/cmd/filter.go	47.27%	22 Missing and 7 partials ⚠️
cmd/skaffold/app/cmd/lsp.go	28.12%	23 Missing ⚠️
cmd/skaffold/app/cmd/verify.go	23.33%	23 Missing ⚠️
cmd/skaffold/app/cmd/fix.go	51.16%	17 Missing and 4 partials ⚠️
cmd/skaffold/app/cmd/inspect_job_manifest_paths.go	60.00%	15 Missing and 1 partial ⚠️
cmd/skaffold/app/cmd/inspect_namespaces.go	50.00%	13 Missing and 1 partial ⚠️
...md/skaffold/app/cmd/inspect_config_dependencies.go	45.83%	12 Missing and 1 partial ⚠️
cmd/skaffold/app/cmd/lint.go	42.85%	12 Missing ⚠️
cmd/skaffold/app/cmd/inspect_build_env.go	60.71%	11 Missing ⚠️
... and 21 more

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #9136      +/-   ##
==========================================
- Coverage   70.48%   63.71%   -6.77%     
==========================================
  Files         515      630     +115     
  Lines       23150    32447    +9297     
==========================================
+ Hits        16317    20674    +4357     
- Misses       5776    10183    +4407     
- Partials     1057     1590     +533     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[renzodavid9]

Hey @beastpu! Thanks for opening this PR. I left a comment for one more place we need to update the resource lists.

Checking the change, I think I found a case where still we'll not be able to get the logs: when the CRD uses a field different than image to define the image to deploy. When we are going through the manifests to get the images, Skaffold is looking for the specific image field:

skaffold/pkg/skaffold/kubernetes/manifest/images.go

Lines 114 to 117 in 1f591ba

	func (is *imageSaver) Visit(gk apimachinery.GroupKind, navpath string, o map[string]interface{}, k string, v interface{}, rs ResourceSelector) bool {
	if k != imageField {
	return true
	}

I think the proper way here would be to use the selectors defined in the resourceSelector from the skaffold.yaml.

With that said, I'll create another issue to track that case and tackle that in another PR; I think we can merge this PR once we have the last change from the comment I left. Thanks a lot for your help! 😄