Fix ClusterDisruptionIT.testAckedIndexing

[henningandersen]

Use assertBusy when doing reroute after bridged disruption,
since it can return non-acked if a node is marked faulty
by follower check after disruption ended.

Closes #53064

[elasticmachine]

Pinging @elastic/es-distributed (:Distributed/CRUD)

[dnhatn]

LGTM, thanks Henning.

[henningandersen]

Thanks for reviewing, Nhat.

[romseygeek]

This test has also failed in 6.8, see https://gradle-enterprise.elastic.co/s/t6xoiqr4ngnp4

Should we backport the fix there as well?

[henningandersen]

@romseygeek , unfortunately that test failure is in a different place:

   > Expected: <true>
   >      but: was <false>
   >    at __randomizedtesting.SeedInfo.seed([74F3D491E0E2587B:FE326062BC9FBE30]:0)
   >    at org.hamcrest.MatcherAssert.assertThat(MatcherAssert.java:20)
   >    at org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertAcked(ElasticsearchAssertions.java:109)
   >    at org.elasticsearch.test.hamcrest.ElasticsearchAssertions.assertAcked(ElasticsearchAssertions.java:113)
   >    at org.elasticsearch.test.TestCluster.wipeIndices(TestCluster.java:142)
   >    at org.elasticsearch.test.TestCluster.wipe(TestCluster.java:79)
   >    at org.elasticsearch.test.ESIntegTestCase.afterInternal(ESIntegTestCase.java:592)
   >    at org.elasticsearch.test.ESIntegTestCase.cleanUpCluster(ESIntegTestCase.java:2205)
   >    at java.lang.Thread.run(Thread.java:748)

I find it reasonable to ignore this one incident since so much has happened in 7.x w.r.t. cluster healing (new coordination layer for instance) that it hardly makes sense to focus on this in 6.x. I would opt to leave the test unmuted in 6.x, unless this reoccurs too frequently (in case it fails in other interesting ways).