we skip onionskins that came from non-relays, so we're less likely to
run into privacy troubles.

starts to implement ticket 9658.

Now we explicitly check for overflow.

This approach seemed smarter than a cascade of "change int to unsigned
int and hope nothing breaks right before the release".

Nick, feel free to fix in a better way, maybe in master.

with commit c6f1668d we let it grow arbitrarily large.

it can still overflow, but the damage is very small now.

Conflicts:
	src/test/test.c

Now we consider the TAP cells we'll process while draining the NTor
queue, and vice versa.

that way tap won't starve entirely, but we'll still handle ntor requests
quicker.

Now we prioritize ntor create cells over tap create cells.

Starts to address ticket 9574.

Conflicts:
	src/or/circuituse.c

Conflicts:
	src/or/circuitbuild.c

This would make us do testing circuits "even when cbt is disabled by
consensus, or when we're a directory authority, or when we've failed
to write cbt history to our state file lately." (Roger's words.)

This is a fix for 9671 and an improvement in our fix for 5049.
The original misbehavior was in 0.2.2.14-alpha; the incomplete
fix was in 0.2.3.17-beta.

Fix for bug 9400, spotted by coverity. Bug introduced in revision 2cb4f7a4
(subversion revision r389).

(In practice they don't exist, but so long as we're making changes for
standards compliance...)

Also add several more unit tests for good and bad URL types.

MyFamily option will now fix fingerprints missing their leading  instead of complaining. Should fix #4341.