PrometheusSpec configuration for Target Allocator

[swiatekm]

Target Allocator can generate Prometheus configurations based on prometheus-operator Custom Resources, and it does so using code imported from prometheus-operator. In order to generate this configuration, the monitoring CRs like ServiceMonitor aren't sufficient, as some of the configuration lives in the Prometheus CR. The default scrape interval is a good example, see #1925.

We should decide which fields from the Prometheus CR we want to support in the Target Allocator spec. There's quite a few fields in https://github.com/prometheus-operator/prometheus-operator/blob/main/Documentation/api.md#commonprometheusfields, and most of them either don't apply to Target Allocator, or do apply, but are already set in the Collector spec.

The purpose of this issue is to track which fields we should add, and their implementation status.

[swiatekm]

An initial list of what seems applicable:

• serviceMonitorSelector
• serviceMonitorNamespaceSelector
• podMonitorSelector
• podMonitorNamespaceSelector
• scrapeTimeout
• overrideHonorLabels
• overrideHonorTimestamps
• ignoreNamespaceSelectors
• enforcedNamespaceLabel
• enforcedSampleLimit
• enforcedTargetLimit
• enforcedLabelLimit
• enforcedLabelNameLengthLimit
• enforcedLabelValueLengthLimit
• enforcedBodySizeLimit
• minReadySeconds
• excludedFromEnforcement
• podTargetLabels

[jaronoff97]

This is something we should do as part of our v2 efforts.