Move the predictive vary from the req->ws to busyobj->ws.

predictive vary could do.

XXX: doc-change needed.

there.  These bits are legitimate feature-ettes which might make sense
in a production environment.

XXX: need doc-change

XXX: doc-update needed

XXX: doc-update needed

XXX: doc-update needed

XXX: doc-update needed

XXX: doc-update needed

suppressed so the never take up space/bandwidth in the VSL.

Reimplement vcl_trace using this:

	param.set vsl_mask +VCL_trace
	param.set vsl_mask -VCL_trace

XXX: doc-update needed

Some more text describing the behaviour on the various storage backend
+ some reformatting for readability

to like it...

Patch by:	daghf

assert before the parent carried out the assignment of the thread
id to the variable.

Detected by:	OSX

Spotted by: OSX pthread implementation

And yes, why ptrdiff_t even exists and why it can be different,
semantically or otherwise, from ssize_t is beyond me...

from sphinx. It was never properly maintained.

for ESI processing.

Polish the VGZ code a bit while here anyway.

Fixes	#1184