Surface removed LCP element in report

[patrickhulce]

Summary
Most of this discussion is happening in chat, but filing an issue so we don't forget. ref #11625 (comment)

LCP changed in m88 to now include removed elements. When this happens and there's a GC we can't find the LCP element anymore and the audit gets marked as not applicable. At that time we may also optionally consider removing _maxChromiumMilestone support from smoke tests.

[paulirish]

in the future there may be more dedicated protocol support: events issued with node ids as they happen.

for now, we have to work around this ourselves:

• set up a perfObserver
  • perhaps just having it with a reference prevents the node from being GC'd
  • if not, then we'll need to report it back from there in ~realtime.

[paulirish]

1138441 - Add Core Web Vitals metrics to Performance.GetMetrics results - chromium tracks the protocol support

[adamraine]

Support was added but this boi is ancient? Do we still want this?

[paulirish]

LargestImagePaint::Candidate is available and gives us image URL and coordinates. But doesn't help on the GC front.

The new CDP events in the above crbug give us the nodes at runtime. We can collect their nodeDetails as those events come in to ensure they're not GC'd before the TraceElements gatherer runs. For this to be worth it, we'd probably have to collect these nodeDetails right when the events come in. But... overhead is unknown and that may be unwise.

Something to look into, especially if we see this problem happening often.