CI Failure (ERROR log from aborted Kafka conn) in ManyPartitionsTest.test_many_partitions

[VladLazar]

https://buildkite.com/redpanda/vtools/builds/9042#018a1c48-0769-4882-9aff-913e6901cd38

Module: rptest.scale_tests.many_partitions_test
Class:  ManyPartitionsTest
Method: test_many_partitions

====================================================================================================
test_id:    rptest.scale_tests.many_partitions_test.ManyPartitionsTest.test_many_partitions
status:     FAIL
run time:   13 minutes 13.255 seconds


    <BadLogLines nodes=ip-172-31-4-173(28),ip-172-31-13-207(44),ip-172-31-3-247(47),ip-172-31-14-108(16),ip-172-31-9-205(20),ip-172-31-4-190(3),ip-172-31-14-231(60),ip-172-31-15-158(36),ip-172-31-3-202(35) example="ERROR 2023-08-22 08:23:16,721 [shard 1:main] kafka - server.cc:154 - Error[applying protocol] remote address: 172.31.13.99:57214 - seastar::abort_requested_exception (abort requested)">
Traceback (most recent call last):
  File "/usr/local/lib/python3.10/dist-packages/ducktape/tests/runner_client.py", line 135, in run
    data = self.run_test()
  File "/usr/local/lib/python3.10/dist-packages/ducktape/tests/runner_client.py", line 227, in run_test
    return self.test_context.function(self.test)
  File "/home/ubuntu/redpanda/tests/rptest/services/cluster.py", line 142, in wrapped
    redpanda.raise_on_bad_logs(
  File "/home/ubuntu/redpanda/tests/rptest/services/redpanda.py", line 1233, in raise_on_bad_logs
    raise BadLogLines(bad_lines)
rptest.services.utils.BadLogLines: <BadLogLines nodes=ip-172-31-4-173(28),ip-172-31-13-207(44),ip-172-31-3-247(47),ip-172-31-14-108(16),ip-172-31-9-205(20),ip-172-31-4-190(3),ip-172-31-14-231(60),ip-172-31-15-158(36),ip-172-31-3-202(35) example="ERROR 2023-08-22 08:23:16,721 [shard 1:main] kafka - server.cc:154 - Error[applying protocol] remote address: 172.31.13.99:57214 - seastar::abort_requested_exception (abort requested)">

[VladLazar]

This seems related to #12021.
When the client disconnects in Kafka::server::apply we set the log and re-throw the exception.
This bubbles up to net::server::apply_proto where we log the exceptional future at error level.

We should special case the client disconnect and log it and different severity level as this is expected in
this test. Alternatively, we could just allow list the log. (cc. @BenPope)

[BenPope]

There's some work on it here: #12793, but I think something is still missing.

I've rebased it and added a small fix here: #12955