refactor(trie): rename persistRoot to useRootPersistence

[faustbrian]

Leaving as draft until #2215 is merged

Bringing this property in line with the naming of other functional modifiers. Also see #2226.

[codecov]

Codecov Report

Merging #2223 (1c5fe9f) into master (fffe4ab) will increase coverage by 0.18%.
The diff coverage is 100.00%.

❗ Current head 1c5fe9f differs from pull request most recent head 4d14f36. Consider uploading reports for the commit 4d14f36 to get more accurate results

Flag	Coverage Δ
block	92.98% <ø> (ø)
blockchain	88.88% <ø> (?)
client	87.00% <ø> (ø)
common	59.55% <ø> (ø)
devp2p	92.33% <ø> (ø)
ethash	∅ <ø> (∅)
evm	79.07% <ø> (ø)
rlp	∅ <ø> (∅)
statemanager	88.15% <ø> (ø)
trie	89.95% <100.00%> (ø)
tx	97.98% <ø> (ø)
util	92.33% <ø> (ø)
vm	85.24% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

[faustbrian]

@holgerd77 conflicts resolved

[faustbrian]

Just ran this against my app and its test suite and all good.

[faustbrian]

@holgerd77 this is green

[faustbrian]

@holgerd77 no conflicts with the checkpoint PR so could admin merge

[holgerd77]

This looks good as well (no approval yet).

[holgerd77]

Hmm, everything is blocked here. Maybe give this combined-PR-suggestion another thought, I just simply otherwise can't guarantee that we still include these changes in the release since I might not have the time any more to approve one-by-one with an hour or so in-between-wait-time. This is just our CI very very much at its limits, since a single merge is triggering re-runs all over the place, not necessarily the right ones queued first.

I am not seeing so much the revert difficulty problem. All things remain in single commits and if one would run at least the Trie tests after each cherry-pick this should remain under control I guess.

So maybe give this another thought, also important to close the taken-over PRs.

[holgerd77]

(and things have mostly already proven to be working by CI having passed on close to all (all?) separate PRs)