rec: if the full CNAME chain leading to the answer is cached, indicate that

[omoerbeek]

Alternative approach to PR #14918 from @zjs604381586.

I find the approach to be more clear. Once we agree which one is best, I'll add a unit test.

Short description

Checklist

I have:

• read the CONTRIBUTING.md document
• compiled this code
• tested this code
• included documentation (including possible behaviour changes)
• documented the code
• added or modified regression test(s)
• added or modified unit test(s)

[coveralls]

Pull Request Test Coverage Report for Build 12370453034

Details

• 109 of 120 (90.83%) changed or added relevant lines in 2 files are covered.
• 48 unchanged lines in 13 files lost coverage.
• Overall coverage increased (+3.6%) to 64.817%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
pdns/recursordist/test-syncres_cc1.cc	88	92	95.65%
pdns/recursordist/syncres.cc	21	28	75.0%

Files with Coverage Reduction	New Missed Lines	%
pdns/ueberbackend.cc	1	82.03%
modules/gpgsqlbackend/gpgsqlbackend.cc	1	88.62%
pdns/recursordist/syncres.cc	1	80.21%
pdns/recursordist/aggressive_nsec.cc	2	66.32%
modules/lmdbbackend/lmdbbackend.cc	2	72.54%
pdns/misc.cc	2	64.57%
pdns/misc.hh	3	87.62%
modules/gpgsqlbackend/spgsql.cc	3	67.94%
pdns/dnsdistdist/dnsdist.cc	4	68.93%
pdns/recursordist/test-syncres_cc1.cc	5	89.95%

Totals
Change from base Build 12369518911:	3.6%
Covered Lines:	126108
Relevant Lines:	163770

---

💛 - Coveralls

[rgacogne]

Looks nice, I would love a test :)

[zjs604381586]

This solution looks better than mine.
For example, if there are multiple layers of cname cache for the domain name in the cache, and the last cname has an A record or the last cname has no record cache, if the AAAA type is requested at this time, it looks like in step 0 cache lookup, only the AAAA type of the last cname is resolved through the network, and the others will be returned through the cache. You can test it.