c/migration/table: when applying a snapshot update completed timestamp

[bashtanov]

When applying a snapshot we forgot to update completed timestamp. It may bring us into an incorrect state where state is finished but completed timestamp is not set.
https://redpandadata.atlassian.net/browse/CORE-8723

Backports Required

• none - not a bug fix
• none - this is a backport
• none - issue does not exist in previous branches
• none - papercut/not impactful enough to backport
• v24.3.x
• v24.2.x
• v24.1.x

Release Notes

• none

[vbotbuildovich]

Retry command for Build#60352

please wait until all jobs are finished before running the slash command

/ci-repeat 1
tests/rptest/tests/archival_test.py::ArchivalTest.test_single_partition_leadership_transfer@{"cloud_storage_type":2}

[bashtanov]

^ irrelevant

[bashtanov]

/ci-repeat 1
tests/rptest/tests/archival_test.py::ArchivalTest.test_single_partition_leadership_transfer@{"cloud_storage_type":2}

[vbotbuildovich]

CI test results

test results on build#60367

test_id	test_kind	job_url	test_status	passed
rm_stm_tests_rpunit.rm_stm_tests_rpunit	unit	https://buildkite.com/redpanda/redpanda/builds/60367#01944251-cd70-40ee-bef7-453555e2b969	FLAKY	1/2
storage_e2e_single_thread_rpunit.storage_e2e_single_thread_rpunit	unit	https://buildkite.com/redpanda/redpanda/builds/60367#01944251-cd70-40ee-bef7-453555e2b969	FLAKY	1/2

[vbotbuildovich]

/backport v24.3.x

[dotnwat]

does this have implications on existing clusters that aren't expecting this new value to appear?

[bashtanov]

It normally appears, as raft message application does transfer it. It may be missed when a snapshot is applied. Whether it comes via message or via snapshot depends on raft lifecycle such as node shutdowns/resumes/recovery etc. So I doubt anything expects it to be missing.