catch trouble that materializes only when we destroy the VGZ instance.

Fixes	#1037
Fixes	#1043
Fixes	#1044

Conflicts:

	bin/varnishd/cache.h
	bin/varnishd/cache_center.c
	bin/varnishd/cache_esi_deliver.c
	bin/varnishd/cache_esi_fetch.c
	bin/varnishd/cache_fetch.c
	bin/varnishd/cache_gzip.c
	bin/varnishd/cache_response.c

to fail, and act accordingly.

Conflicts:

	bin/varnishd/cache_fetch.c

to clean those up non-fatal.

Fixes	#1036

Conflicts:

	bin/varnishd/cache_gzip.c

The thread (s%d/c%d) still dies, but it does not mark this as
fatal with respect to the test as such.

This is useful where varnishd shuts the connection on a backend
prematurely, as part of the correct execution of a test.

Fix a undue recursion problem with asserts in the vtc_log code.

before all diagnostics have been gathered.

Conflicts:

	bin/varnishtest/vtc.c

processing.

Convert a few http_PrintfHeader() to http_SetHeader() for good
measure: There is no reason to waste workspace on compiled in strings.

Fixes	#1038

Conflicts:

	bin/varnishd/cache_center.c
	bin/varnishd/cache_response.c

Remove reference/purging_banning, since all info is now in tutorial/purging

Submitted by:	Nils Goroll

crashing child to do so, and then explicitly start it again.

avoid fixed sleeps waiting for the child process to start.

1 thread is a bit on the low side, make the default on RPM based
distros 50.

instead of asserting.

Inspired by:	Patch from DocWilco

fd's early in the case of a timeout.

Fixes	#1023

Conflicts:

	bin/varnishd/cache_waiter_poll.c

Reverting 6 commits seems like more confusion. This is varnishncsa.c from d18336e5.