this is a bad leftover from when we still used ALLOC_JOB/FREE_OBJ for
nodes and anyway we must not just free a node which is still linked in
an otherwise alive tree

During code restructuring earlier in
77e9559d I overlooked that VTIM_real()
may not be declared where debug output is used

solve by moving the accumulation to the destructor

it is in the way of my torture testing and does not smell like something
we have caused..

this avoids a potential panic for an out of workspace condition.

Also, we move error handling to a place where it is much simpler

much remains to be done

ved_ is used in varnish-cache

there also was a stupid precedence error in PF_CFG_DEFAULT :(

docs are still WIP

Remove the increased thread_pool_min setting where it's not needed,
and fix indentation.

we now get the flags from vcl once when we init the pesi vdp

this should preserve existing functionality, but is not complete with
regards to the configuration from vcl.