do not enforce 1/3 validator power change

[p2p] Pong Timeout

leave it to the app

Refs #1022

* extract stopPongTimer method
* TestMConnectionMultiplePings

Benchmark results:

```
BenchmarkSwitchBroadcast-2         30000             71275 ns/op
--- BENCH: BenchmarkSwitchBroadcast-2
        switch_test.go:339: success: 1, failure: 0
        switch_test.go:339: success: 100, failure: 0
        switch_test.go:339: success: 10000, failure: 0
        switch_test.go:339: success: 30000, failure: 0
```

Update tmlibs & protobuf deps

improve "curRate too low" message

See https://github.com/tendermint/tendermint/issues/1177#issuecomment-363720118

Refs #1177

Note on labels:
KB - 1024
kB - 1000
https://ux.stackexchange.com/questions/13815/files-size-units-kib-vs-kb-vs-kb

WIP: TestReactorValidatorSetChanges fails non deterministically

fix memory leak in mempool reactor

add seed_mode flag (`--p2p.seed_mode`)

With must be called on log.filter, otherwise "main" entries get filtered

types: TxEventBuffer.Flush now uses capacity preserving slice clearing idiom

Distribution improvements (freeze glide version + get rid of gox)

don't bother with this "only ping when we havent heard from them". lets
just always ping every peer from the sendRoutine every 10s no matter
what. if they dont pong within pongTimeout, disconnect :)

Also, we should allow "main" module to log INFO messages like

```
I[02-07|07:57:25.074] Found private validator                      module=main path=/home/vagrant/.tendermint/config/priv_validator.json
I[02-07|07:57:25.076] Found genesis file                           module=main path=/home/vagrant/.tendermint/config/genesis.json
```

Refs https://github.com/cosmos/gaia/issues/118

**BEFORE**:
```
$ tendermint init

```

**AFTER**:
```
$ tendermint init
I[02-07|07:57:25.074] Found private validator                      module=main path=/home/vagrant/.tendermint/config/priv_validator.json
I[02-07|07:57:25.076] Found genesis file                           module=main path=/home/vagrant/.tendermint/config/genesis.json
```

Update getting-started.rst with Python 3 example

get rid of gox

build target builds inside docker, dev-build - locally

Revert "build target builds inside docker, dev-build - locally"

This reverts commit 8ba89d5e8c.

add build tags to make build/build_race/install

use tendermint's fork of glide instead of tar.gz

remove TMHOME unused var + set length for git hash

get rid of GOTOOLS_CHECK

fixes after review

zip

needed for distribution

Fixes https://github.com/tendermint/tendermint/issues/1189

For every TxEventBuffer.Flush() invoking, we were invoking
a:
  b.events = make([]EventDataTx, 0, b.capacity)
whose intention is to innocently clear the events slice but
maintain the underlying capacity.

However, unfortunately this is memory and garbage collection intensive
which is linear in the number of events added. If an attack had access
to our code somehow, invoking .Flush() in tight loops would be a sure
way to cause huge GC pressure, and say if they added about 1e9
events maliciously, every Flush() would take at least 3.2seconds
which is enough to now control our application.

The new using of the capacity preserving slice clearing idiom
takes a constant time regardless of the number of elements with zero
allocations so we are killing many birds with one stone i.e
  b.events = b.events[:0]

For benchmarking results, please see
https://gist.github.com/odeke-em/532c14ab67d71c9c0b95518a7a526058
for a reference on how things can get out of hand easily.

https://github.com/tendermint/tendermint/pull/1173#issuecomment-363173047

https://play.golang.org/p/gN21yO9IRs3

```
func waitWithCancel(f func() *clist.CElement, ctx context.Context) *clist.CElement {
	el := make(chan *clist.CElement, 1)
	select {
	case el <- f():
```
will just run f() blockingly, so this doesn't change much in terms of behavior.

Leaking goroutine:
```
114 @ 0x42f2bc 0x42f3ae 0x440794 0x4403b9 0x468002 0x9fe32d 0x9ff78f 0xa025ed 0x45e571
```

Explanation:
it blocks on an empty clist forever. so unless theres txs coming in,
this go routine will just sit there, holding onto the peer too.
if we're constantly reconnecting to some peer, old instances are not
garbage collected, leading to memory leak.

Fixes https://github.com/cosmos/gaia/issues/108
Previous attempt https://github.com/tendermint/tendermint/pull/1156