security: Rate limit GetAddr responses

[arya2]

Motivation

We want to avoid giving out Zebra's entire address book over a short duration.

Closes #7823.

PR Author Checklist

Check before marking the PR as ready for review:

• Will the PR name make sense to users?
• Does the PR have a priority label?
• Have you added or updated tests?
• Is the documentation up to date?

For significant changes:

• Is there a summary in the CHANGELOG?
• Can these changes be split into multiple PRs?

If a checkbox isn't relevant to the PR, mark it as done.

Solution

• Updates ADDR_RESPONSE_LIMIT_DENOMINATOR to 4
• Adds a 10 minute INBOUND_CACHED_ADDRS_REFRESH_INTERVAL constant
• Adds a CachedPeerAddrs struct with a reference to the address book
• Replaces the address_book field in Setup::Initialized with an instance of CachedPeerAddrs
• Refreshes the cached addresses in the inbound service's poll_ready() method if they are stale
• Returns cached_addrs in response to inbound::Peers requests instead of reading from the address book
• Logs a warning if it hasn't been able to get a lock on the address book to update the cached_addrs in an unexpectedly long time

Related cleanups:

• Moves logic for truncating list of peers for GetAddr response to a sanitized_window() method on AddressBook.

Testing

TODO: This PR still needs tests.

Review

Anyone can review.

Reviewer Checklist

Check before approving the PR:

• Does the PR scope match the ticket?
• Are there enough tests to make sure it works? Do the tests cover the PR motivation?
• Are all the PR blockers dealt with?
  PR blockers can be dealt with in new tickets or PRs.

And check the PR Author checklist is complete.

[teor2345]

This looks good, I think it would be enough to just test that we get peers when we send an address message.

[teor2345]

I think we might be over-complicating this change a bit. Just change the constant?

[teor2345]

Nothing here is blocking, so we're good to go when you're happy with the remaining test and module changes.

[teor2345]

From the CI test results, it looks like some more test or production updates are needed.

[arya2]

From the CI test results, it looks like some more test or production updates are needed.

Yeah, that real_peer_set test seems to be broken in CI (but works locally).

I think this needs another test too.

[teor2345]

The docker CI failure is #7898

[arya2]

I used up the Github actions limit with frequent commits. 😞

[teor2345]

Thanks for all the changes, this looks good for now