The default is for connections to last forever[1]. This leads to fds
leaking. I set the timeout so high by default so that watches don't have
to keep retrying but perhaps we should set it slower.
Tested on a cluster with lots of clients and it seems to have relieved
the problem.
[1] https://groups.google.com/forum/#!msg/golang-nuts/JFhGwh1q9xU/heh4J8pul3QJ
In d0c4916fe9b2afaa273a2a7bc9782321a866ab9f the TLS CA Certificate
verification broke.
This was bisected using the following basic test:
```
./bin/etcd -f -name machine0 -data-dir machine0 -ca-file=/tmp/ca/ca.crt -cert-file=/tmp/ca/server.crt -key-file=/tmp/ca/server.key.insecure
```
And in another window doing
```
curl --key /tmp/ca/server2.key.insecure --cert /tmp/ca/server2.crt -k -L https://127.0.0.1:4001/v2/keys/foo -XPUT -d value=bar -v
```
Before merging this PR there are a few things that need to be fixed up:
1) Tests for client certs both positive and negative
2) Refactor (or at least documentation of) the TLSConfig types
