Split the worker into a distribution of its own

A cluster or GPU host installs `pip install fluksio-worker` and gets the
agent and the runner, not psycopg, numpy and the MCP SDK. The engine
depends on it as a workspace member, so the file it launches node code
with is the same file a remote worker runs — which is what keeps a node
unable to tell the difference. Copying the two files by hand still works.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>

worker/fluksio_worker/__init__.py

@@ -0,0 +1,6 @@
+"""The Fluksio worker: node execution on a machine the engine cannot reach.
+
+Deliberately import-free. The engine imports :mod:`fluksio_worker.worker_main`
+to find the runner it launches as a subprocess, and paying for the agent's
+imports on that path would be a cost for nothing.
+"""