Debrief for 2020.04

Summary

All of the identified high priority PRs were included. It's important to be realistic with this.
Start running a subset of spec tests some weeks before the freeze helps to find some bugs on early stages.
During testing automation has shown again to help a lot => RIOTnode should be pushed.
Knowing area maintainers helps when issues pop up.
Having the milestone set in PRs helps => Let's mark them when merging.

The Managing a Release guide was SUPER useful. Get really familiar with it with time.
Write down advance searches from GitHub which you will have to make several times during the release process.
Asking for high-priority PRs seemed good. We got all of them in. But take a look on the state before, and be realistic when adding them to the list.
Run as many tests as you can before the freezes.
Start tagging and adding the milestone to already merged PRs with time. Iterate over this every a couple of days. Followed @fjmolinas advice "tag as minor stuff you wouldn't put in the Release notes". It helps to get acquainted with the ongoing work.

Dates:
- soft freeze: 1st... but as @fjmolinas pointed out in last release it was not so important
- hard freeze: 9th, take holidays into account
- release on the last day of the month
There was no hack 'n ack before the freeze in this case, because of the COVID-19 situation.
Line between soft and hard freeze should be evaluated case by case. E.g. we got #13349 in the same day as the hard feature freeze was starting.
We don't tend to set the impact: major label

Automation really helps. We should get RIOTnode in for 2020.07.
Tests with clear guides (if step-by-step even better) were much easier to run for the first time.
Not many maintainers were active during release testing this time.
Almost everything was OK in RC1 already.
It's useful to do a summary of every RC (tests that should be re-run, merged PRs, etc.) as the first thing for the next one.
Knowing area maintainers helps when issues pop up.