the global meter.

features

Ensures that all raw data has at least an alignment of 128.

We don't want to link to pythonlib when compiling the dora-daemon.

Buffer capacity is always 0 if the buffer is allocated externally. Also, we don't want to allocate unneccessary extra space if only parts of a buffer a used (e.g. when slicing).

This PR/commit adds documentation for the send_output method.

Follow up discussion: https://github.com/orgs/dora-rs/discussions/317#discussioncomment-6475816

add an attribute that disables the dead_code lint

Previous implementation of the runtime would link `libpython` even though it
might not be required.

This commit put pyo3 dependency under a feature flag that is only active if
the runtime is called from python for a python operator.

This fix:
- https://github.com/dora-rs/dora/actions/runs/4879188973/jobs/8705515472
- https://github.com/dora-rs/dora/issues/280#issuecomment-1534058207
- and https://github.com/dora-rs/dora-drives/issues/58#event-9162532660

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.

In #229, I renamed `eyre::eyre!("{err}{traceback}")`  where it should have `eyre::eyre!("{traceback}\n{err}")` . This PR fix this issue.

This Pull Request also add the name of the node process at the end of the process call to help with debugging.

Instead of doing an additional copy to send them from the operator thread to the runtime thread.

This commit uses the new `allocate_data_sample` and `send_output_sample` methods introduced in d7cd370.

This is mostly about convenience when working with arrow data types, as the data is still being copied once.

Reuse same event type as for Python node API.

The previous approach no longer works on `pyo3` v0.18 because `setattr` now requires a `IntoPy<Py<PyString>>` as key.