Fixes #2871.

Conflicts:
	bin/varnishd/cache/cache_fetch.c

Submitted by:   Frederic Lecaille <flecaille@haproxy.com>

It turns out people are really attached to their varnish.params file.

See `git grep PROXY -- bin/varnishd/mgt/mgt_main.c`.

A regression from d7a81fe8 and the
followup changes from #2993. The new VSS_Resolve{One,First} functions
expect a non-null address string.

Conflicts:
	lib/libvmod_std/vmod_std_conversions.c

Conflicts:
	lib/libvmod_std/vmod.vcc

Conflicts:
	bin/varnishtest/tests/m00011.vtc
	lib/libvmod_debug/vmod.vcc
	lib/libvmod_debug/vmod_debug.c

And sync the documentation with the current behavior, part of which used
to be implicit.

Conflicts:
	lib/libvmod_std/vmod.vcc
	lib/libvmod_std/vmod_std_conversions.c

Conflicts:
	lib/libvmod_std/vmod_std_conversions.c

This is how std.ip is documented, so VSS_ResolveOne doesn't work there.
It might not be the only migration to VSS_ResolveOne that requires
attention.

Speaking of attention, VSA_Malloc may require some.

Conflicts:
	lib/libvmod_std/vmod_std_conversions.c

It can now be shared by established callback-based resolvers and the new
VSS_ResolveOne. This also changes the semantics of VSS_ResolveOne in the
sense that the port is now a default port, overriden by the address if
it contains one.

Also make it clear that VTCP was already relying on a VSS function that
didn't and still doesn't allow a null errp argument, while conversely
all VTCP_open call sites provide a valid errp argument.

At this point the workspace already overflowed so there's no point
making further progress to fail on the next workspace operation.

Incidentally, we are failing for an IP conversion, not an integer.

Conflicts:
	lib/libvmod_std/vmod_std_conversions.c

we want a single unique suckaddr.

Conflicts:
	bin/varnishtest/vtc_client.c
	lib/libvmod_std/vmod_std_conversions.c

not need it (acceptor, h2_sess)

The moment we unlock the oh, another thread could rush req

Partially undoes f35b86bc

Huge thank you to @mbgrydeland for spotting my mistake

and polish the output formatting by reordering

Ref #1799

Ref #2743

see 11 lines down

Conflicts:
	bin/varnishd/cache/cache_hash.c

req's on waiting list do indeed hold an objhead ref (req->hash_objhead),
so the condition for the while loop in hsh_deref_objhead_unlock will
always evaluate to false.

This gets rid of the needless rushing code and replaces it with an
assert.

Simplify the !busy cases

Inspired by:	#2856

Use req->hash_always_miss directly

Inspired by:    #2856

Refactor the hit-cases

Inspired by:	#2856

Move successful loop termination out of the loop.

Inspired by:	#2856