`pytest --dist=loadgroup` hangs if a crashed worker is restarted

[radoering]

When running the following test file

def test_1():
    assert True

def test_2():
    import time
    time.sleep(5)
    assert True

with (uses features from pytest 9)

pytest -n1 --dist=loadgroup -o faulthandler_timeout=1 -o faulthandler_exit_on_timeout=true testing/test_timeout.py

test execution hangs with

replacing crashed worker gw0
collecting: 1/2 workers
collecting: 1/2 workers
2 workers [2 items]

When using another test distribution algorithm, test execution does not hang.

The issue is that

pytest-xdist/src/xdist/scheduler/loadscope.py

Lines 199 to 200 in 8fed345

	# Made uncompleted work unit available again
	self.workqueue.update(workload)

adds the complete workload to the queue, including the completed work.

Later in

pytest-xdist/src/xdist/scheduler/loadscope.py

Lines 276 to 280 in 8fed345

	nodeids_indexes = [
	worker_collection.index(nodeid)
	for nodeid, completed in work_unit.items()
	if not completed
	]

completed work items are dropped so that nodeids_indexes is empty.

Finally, the process hangs in

pytest-xdist/src/xdist/remote.py

Line 204 in 8fed345

self.nextitem_index = self.torun.get()