I added a test to confirm the output matches the old.

Fixes clippy warning clippy::format_collect
https://rust-lang.github.io/rust-clippy/master/index.html#/format_collect

We are not interested in the local bind address of the daemon. Instead, we want to use the IP address under which the daemon is available from other machines.

This should also avoids the issue that connecting to 0.0.0.0 is not possible on Windows (we want to use 0.0.0.0 as default bind address).

This reverts commit 1ba763cd67.

Ref: #459

Tokio background tasks are canceled when the program exits, so we need some additional synchronization to ensure that the `DestroyResult` is actually sent out.

Make dataflow descriptor available to Python nodes and operators

The dataflow might be finished already when `dora stop` is invoked, which caused an error previously. This commit changes this behavior to return the dataflow result instead in this case.

The name generator uses a thread-local random number generator internally, which makes it `!Send`. So we should not keep it across await points because our `multiple_daemons` example needs to spawn it in the tokio runtime, which requires `Send`.

The previous watchdog requests required waiting for a reply, which could slow down the system under load and also lead to false errors on slower systems (e.g. the CI). Because of this, this commit replaces the watchdog requests with asynchronous heartbeat messages, which don't require replies. Instead, we record the last heartbeat timestamp whenever we receive a heartbeat message. We then periodically check the time since the last received heartbeat message and consider the connection closed if too much time has passed.

In previous iteration of logging, the logging thread was not waited for
and the daemon would exit before finishing to log all lines causing a panic.

See: https://github.com/dora-rs/dora/actions/runs/4805449241/jobs/8551873559#step:11:619

We synchronize the start of nodes (even across machines) to avoid missed messages. This led to deadlock issues when nodes exited before they initialized the connection to the dora daemon. The reason was that the other nodes were still waiting for the stopped node to become ready.

This commit fixes this issue by properly handling node exits that occur before sending the subscribe message. If nodes exit, they are removed from the pending list and added to a new  'exited before init' list. Once all nodes have subscribed or exited, we answer the pending subscribe requests. If nodes exited before subscribing we send an error reply to the other nodes because a synchronized start is no longer possible.

To make this logic work across different machines too, we add a `success` field to the `AllNodesReady` message that the daemon sends to the coordinator. The coordinator forwards this flag to other daemons so that they can act accordingly.

The coordinator is our control plane and should not be involved in data plane operations. This way, the dataflow can continue even if the coordinator fails.

We still need to pass the path through a new `working_dir` field as we haven't figured out deployment of compiled nodes/operators yet.

Removes the separate `dora-runtime` binary. The runtime can now be started by passing `--run-dora-runtime` to `dora-daemon`. This change makes setup and deployment easier since it removes one executable that needs to be copied across machines.

Allow passing runtime path via coordinator