[cvas22]

What this PR does / why we need it:

Adds the flushIntervalSec and flushIntervalNsec options to the Fluent Bit logToMetrics filter (ClusterFilter and Filter). They're rendered as Fluent Bit's Flush_Interval_Sec and Flush_Interval_Nsec.

Without them, the filter emits metrics right after every matching record. #1459 reports this produces duplicate samples with the same timestamp and different values in Prometheus. With an interval set, metrics are emitted once per interval, which also lowers resource use under load.

Both fields are optional and validated as non-negative (flushIntervalNsec up to 999999999). When they aren't set nothing is rendered, so existing configs are unchanged.

filters:
  - logToMetrics:
      tag: metrics
      metricMode: counter
      metricName: count_all
      flushIntervalSec: 30

Which issue(s) this PR fixes:

Fixes #1459

Does this PR introduced a user-facing change?

The Fluent Bit logToMetrics filter supports flushIntervalSec and flushIntervalNsec.

Notes

I ran the generators (make generate manifests docs-update, in WSL with controller-gen v0.18.0) rather than editing generated files by hand. They updated the filter deepcopy, the filters CRDs (config, both charts, manifests/setup/setup.yaml) and docs/plugins/fluentbit/filter/log_to_metrics.md.

Tested locally: go test ./apis/... passes, including the new TestFilter_LogToMetrics_* cases (with and without the interval), and make lint reports 0 issues.

[joshuabaird]

LGTM, thanks for this. The changes follow the existing plugin patterns and the generated files look consistent.

One optional nit: it might be worth mentioning the 0-999999999 range in the flushIntervalNsec description so it shows up in the docs.