bootstrap add --default option

this commit:
* moves parsing of bootstrap peers into config
* moves location of bootstrap addrs into core/commands
* refactor `*BootstrapPeer -> BootstrapPeer

sharness bootstrap rm

if a test is cancelled, unmount may not run. then, must
run umount before mounting, just to make sure.

sharness tests should not bootstrap to the dht. it
may be useful to test that specifically, but not on
the other tests.

feat(dht) Bootstrap

use dht bootstrap. there is an edge case where the dht
is tiny (1?) and we have 0 bootstrap peers. we should
probably _inform_ the user, but this may be more a
webui or command thing.

This also makes it an Error to find a peer.

to test jenkins

License: MIT
Signed-off-by: Brian Tiger Chow <brian@perfmode.com>

License: MIT
Signed-off-by: Brian Tiger Chow <brian@perfmode.com>

various fixes

Bitswap doesn't usually care about dialing. the underlying
network adapter can make sure of that.

We use make test as the measure of correctness.
This laxity has let bugs creep into several systems.
This commit changes our target to always run expensive
tests, unless one specifically runs `make test_short`

(we would do well to remove most if not all timing--
that's usually what makes tests take a long time.)

This is causing test failures because tests don't usually
have "/-/-" format. we can decide whether or not to allow
keys without validators, but for now removing.

cc @whyrusleeping

mocknet indeterminism screwed this test up. that's twice
it's bitten us. let's not let it do it a third time.

cc @briantigerchow omg.

TestGetFailures may just be operating very slowly, instead
of completely failing. Right now it gets caught on travis
often. not sure if its actually wrong.

this fixes a failing ipns test which didnt have
a "working" routing system

Without `-f`, `make clean` fails on machines that don't have the dir.

cc @jbenet

@jbenet @whyrusleeping

This bug (missing return) could tie up the client worker and cause
operations to come to a halt.

utp is BROKEN!! it causes tests to fail.

we shouldn't use an arbitrary timeout here. since Get
doesnt take in a context yet, we give a large upper bound.
think of an http request. we want it to go on as long as
the client requests it.