ScyllaDBCluster sync duplication across types

[mflendrich]

This is an extract of the conversation in https://github.com/scylladb/scylla-operator/pull/2524/files#r2017366718 to a separate issue.

The automated multi-region capability implementation has introduced substantial repetition of almost-identical logic across the range of types we sync with the remote cluster:

scylla-operator/pkg/controller/scylladbcluster/sync.go

Lines 223 to 299 in 468cbfa

	err = controllerhelpers.RunSync(
	&status.Conditions,
	remoteNamespaceControllerProgressingCondition,
	remoteNamespaceControllerDegradedCondition,
	sc.Generation,
	func() ([]metav1.Condition, error) {
	return scc.syncNamespaces(ctx, sc, remoteNamespaceMap, managingClusterDomain)
	},
	)
	if err != nil {
	errs = append(errs, fmt.Errorf("can't sync remote namespaces: %w", err))
	}
	err = controllerhelpers.RunSync(
	&status.Conditions,
	remoteRemoteOwnerControllerProgressingCondition,
	remoteRemoteOwnerControllerDegradedCondition,
	sc.Generation,
	func() ([]metav1.Condition, error) {
	return scc.syncRemoteOwners(ctx, sc, remoteNamespaces, remoteRemoteOwnerMap, managingClusterDomain)
	},
	)
	if err != nil {
	errs = append(errs, fmt.Errorf("can't sync remote remoteowners: %w", err))
	}
	err = controllerhelpers.RunSync(
	&status.Conditions,
	remoteServiceControllerProgressingCondition,
	remoteServiceControllerDegradedCondition,
	sc.Generation,
	func() ([]metav1.Condition, error) {
	return scc.syncServices(ctx, sc, remoteNamespaces, remoteControllers, remoteServiceMap, managingClusterDomain)
	},
	)
	if err != nil {
	errs = append(errs, fmt.Errorf("can't sync remote services: %w", err))
	}
	err = controllerhelpers.RunSync(
	&status.Conditions,
	remoteEndpointSliceControllerProgressingCondition,
	remoteEndpointSliceControllerDegradedCondition,
	sc.Generation,
	func() ([]metav1.Condition, error) {
	return scc.syncEndpointSlices(ctx, sc, remoteNamespaces, remoteControllers, remoteEndpointSlicesMap, managingClusterDomain)
	},
	)
	if err != nil {
	errs = append(errs, fmt.Errorf("can't sync remote endpointslices: %w", err))
	}
	err = controllerhelpers.RunSync(
	&status.Conditions,
	remoteEndpointsControllerProgressingCondition,
	remoteEndpointsControllerDegradedCondition,
	sc.Generation,
	func() ([]metav1.Condition, error) {
	return scc.syncEndpoints(ctx, sc, remoteNamespaces, remoteControllers, remoteEndpointsMap, managingClusterDomain)
	},
	)
	if err != nil {
	errs = append(errs, fmt.Errorf("can't sync remote endpoints: %w", err))
	}
	err = controllerhelpers.RunSync(
	&status.Conditions,
	remoteScyllaDBDatacenterControllerProgressingCondition,
	remoteScyllaDBDatacenterControllerDegradedCondition,
	sc.Generation,
	func() ([]metav1.Condition, error) {
	return scc.syncScyllaDBDatacenters(ctx, sc, remoteNamespaces, remoteControllers, remoteScyllaDBDatacenterMap, managingClusterDomain)
	},
	)
	if err != nil {
	errs = append(errs, fmt.Errorf("can't sync remote scylladbdatacenters: %w", err))
	}

plus the secret/CM syncers added in #2524.

The scope of this issue is to come up with some pattern to prevent the sprawl of almost-identical implementations under the ScyllaDBCluster controller.

Acceptance Criteria

• Meaningfully reduced duplication of logic under pkg/controller/scylladbcluster's sync.go.