Proper child table restoration

[puddly]

Aqara sensors specifically are affected by this. When the coordinator receives a Data Request from an unknown child, it sends it a Leave (with rejoin) command. This works to properly initialize the child for everything but older Aqara, which never do the "rejoin" part.

This I think is two SDK bugs:

1. the public child restoration API doesn't work. We have to manually edit NVRAM.
2. the coordinator clearly receives the data, it just does not forward it to bellows. I guess the spec mandates this behavior since the incoming packet is not from a known neighbor or child but we still need to find a way work around this.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 99.46%. Comparing base (0960652) to head (6bdf759).
⚠️ Report is 2 commits behind head on dev.

Additional details and impacted files

@@           Coverage Diff           @@
##              dev     #686   +/-   ##
=======================================
  Coverage   99.45%   99.46%           
=======================================
  Files          58       58           
  Lines        3851     3915   +64     
=======================================
+ Hits         3830     3894   +64     
  Misses         21       21           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.