MongoDB dashboard replication lag count incorrect

I'm gathering metrics via percona/mongodb-exporter and currently trying to understand how replication lag is counting.

When I check `rs.printSlaveReplicationInfo()` on primary - it's not corresponding to the query I've found in MongoDB_Cluster_Summary dashboard https://github.com/percona/grafana-dashboards/blob/master/dashboards/MongoDB_Cluster_Summary.json#L1643

If I change max to min, the value becomes closer to real output of mongo command.

The version of mongodb-exporter is latest (0.8.0)

Thanks in advance.

Steps to reproduce: (cannot reproduce)

Connect to the RS primary and run:

cfg = rs.conf()
cfg.members[1].priority = 0
cfg.members[1].hidden = true
cfg.members[1].slaveDelay = 3600
rs.reconfig(cfg) 

Check the exporter. The lag should be 3600 seconds.

Created draft PR with test