Connection and query timeouts

### Problem

- [ ] If MySQL does not respond as expected, there should be a timeout on the connection and query attempts.
- [ ] Timeout should be set to a reasonable value.
- [ ] User should be able to specify a timeout value.

### Original reported issue

```
gh-mysql-config-diff-apply does not behave well when the MySQL server is running but not behaving correctly.

Details
[hostname] (part of a vitess keyspace), MySQL was in some weird state. It reported the following error message:

Error 1135: Can't create a new thread (errno 11); if you are not out of available memory, you can consult the manual for a possible OS-dependent bug

When I logged into the machine, I saw that there were hundreds of processes related to gh-mysql-config-diff-apply running.

I killed the processes, killed the mysql server, and restarted it. Things seem to be back to normal now.

I'm not sure whether this is a symptom or the cause of MySQL being in a weird state.

Either way, gh-mysql-config-diff-apply needs to be better at handling error conditions and timing out conservatively if MySQL does not behave as expected.
```

Start at the gh-mysql-config-diff-apply entry point and trace how MySQL connections and queries are made. Determine where conservative default timeouts and a user-specified timeout belong; done means both connection and query attempts stop reliably when MySQL does not respond.

Written by the indexing model from the issue text.