chromedp: domEvent: timeout waiting for node

This is more a question than a bug report.

When a dom event listener is about to wait for a node and the node gets concurrently removed (or its parent), WaitNode fails with timeout. The problem here is when that happens, the whole action handlers stale for 10 seconds. At least this is what I think happens. The behaviour is random, if WaitNode finishes before it’s removed, then everything works as expected (I think also Go scheduler contributes to the racy behaviour).

Debug log for timeout scenario:

scrap.go:91: -> {"method":"DOM.childNodeInserted","params":{"parentNodeId":7,"previousNodeId":8,"node":{"nodeId":9,"backendNodeId":9,"nodeType":1,"nodeName":"DIV","localName":"div","nodeValue":"","childNodeCount":0,"attributes":[]}}}
scrap.go:91: -> {"method":"DOM.attributeModified","params":{"nodeId":9,"name":"id","value":"someotherid"}}
scrap.go:91: -> {"method":"DOM.childNodeInserted","params":{"parentNodeId":7,"previousNodeId":0,"node":{"nodeId":11,"backendNodeId":11,"nodeType":1,"nodeName":"BODY","localName":"body","nodeValue":"","childNodeCount":1,"attributes":["style","visibility: hidden; width: 0px; height: 0px; border: 0px; margin: 0px; background: none;"]}}}
scrap.go:91: -> {"method":"DOM.childNodeCountUpdated","params":{"nodeId":11,"childNodeCount":0}}
scrap.go:91: -> {"method":"DOM.childNodeRemoved","params":{"parentNodeId":7,"nodeId":9}}
scrap.go:91: -> {"method":"DOM.childNodeRemoved","params":{"parentNodeId":7,"nodeId":11}}
scrap.go:91: could not perform (attributeModified) operation on node 9 (wait node): timeout waiting for node `9`
scrap.go:91: could not perform (childNodeCountUpdated) operation on node 11 (wait node): timeout waiting for node `11`

I was thinking about fixing this (the 10s stales are really bugging me) by introducing some kind of context cancellation / listening channel, so we break out of waiting loop in WaitNode when the node we’re waiting for gets removed.

Wdyt?