change from fixed to to append new line to slice

[bhpratt]

Community Note

• Please vote on this pull request by adding a 👍 reaction to the original pull request comment to help the community and maintainers prioritize this request
• Please do not leave "+1" or other comments that do not add relevant new information or questions, they generate extra noise for pull request followers and do not help prioritize the request

Fixes issue with older worker pools that do not have os label applied and were hitting out of bounds error (because there were no host labels on the worker pool): panic: runtime error: makeslice: len out of range

Since we cannot know the number of labels, change is to start with a slice of size 0 and increase slice length as number of labels increases.

Output from acceptance testing:

make testacc TEST=./ibm/service/satellite TESTARGS='-run=TestAccSatelliteCluster_Basic'
==> Checking that code complies with gofmt requirements...
TF_ACC=1 go test ./ibm/service/satellite -v -run=TestAccSatelliteCluster_Basic -timeout 700m
--- PASS: TestAccSatelliteCluster_Basic (4573.43s)
PASS
ok      github.com/IBM-Cloud/terraform-provider-ibm/ibm/service/satellite       4574.894s

make testacc TEST=./ibm/service/satellite TESTARGS='-run=TestAccSatelliteClusterWorkerPool_Basic'

==> Checking that code complies with gofmt requirements...
TF_ACC=1 go test ./ibm/service/satellite -v -run=TestAccSatelliteClusterWorkerPool_Basic -timeout 700m
--- PASS: TestAccSatelliteClusterWorkerPool_Basic (4886.39s)
PASS
ok      github.com/IBM-Cloud/terraform-provider-ibm/ibm/service/satellite       4887.689s

I also tested this locally and hit no issues - though I do not have a worker pool w/no host labels to truly replicate the error scenario.