…481) (#498)

Use `serde` directly instead of ask for user of the library to add it as a
dependency.

Tests are modified since this is a ui change.

The doc comment gives an example using `#[rpc(server)]` instead of original
`#[rpc]`, telling users that this attribute has an option to be used to
configure it.

* Add Send + Sync test.

* Add test to make sure the tpc server is Send+Sync too.

* Review suggestions.

* Make sure to use latest tokio version.

* ignore dropped connections

* bump version, warn instead of trace

* added giltab-ci

* removed '--lock' from test

* no time command in onw windows

* added stage: checkstyle

* shortend the config a bit

* bit more compact ci-file

* gitlab-ci intergation

* changed order of ci jobs

* Re-format.

* Disable travis & appveyor.

* Fix fmt.

Manually dropping the future passed from `Future::select` before sending
`done_tx` prevents the race condition.

`Future::select` pass the unresolved future, which is a `Server` holding
`rpc_handler`, to the following callback. Therefore, it is dropped after
the `done_tx.send(())` after the callback exits.
It is possible that the thread that has executed `wait()`, which is
usually the main thread, terminates before the thread sending `done_tx`
drops `rpc_handler`.
Static variables are destructed at the end of termination of the main
thread, and then a segfault occurs when the thread dropping the
rpc_handler accesses the variables.

* Update syn, quote & proc-macro2

* Apply review suggestions.

* Re-format.