Improve chain-reset helm chart

[Pitasi]

After we have zero downtime bulk imports we can have huge improvements to the chain reset helm chart, in particular:

• there's no need to disable the chain and stop all the nodes
• there's no need to erase all the data for that chain

That said, the bulk import still needs to run with a PVC that is not currently being used by a running node (due to LevelDB limitations).

The new helm chart should do something like this (but this is only one possibility):

• stop only one of the replicas of the chain, let's say osmosis-3
• run bulk import against the osmosis-3 PVC
• recreate/start the osmosis-3 pod

Another possibility to be investigated would be (if state-sync is fast enough):

• create a temp pvc
• start a new temp node, let's call it osmosis-temp
• run state-sync and wait for it to populate the pvc
• delete the osmosis-temp pod
• run bulk import against osmosis-temp pvc
• delete osmosis-temp pvc