bug: not able to properly cleanup old partitions

[Ivansete-status]

Problem

waku.sandbox nodes cannot properly delete old partitions.

F.e., the following log appears in waku.sandbox-HK node:

nim-waku  | DBG 2025-08-06 13:38:41.053+00:00 executing message retention policy         topics="waku archive" tid=1 file=archive.nim:254
nim-waku  | DBG 2025-08-06 13:38:41.053+00:00 beginning of executing message retention policy - size topics="waku archive retention_policy" tid=1 file=retention_policy_size.nim:21
nim-waku  | ERR 2025-08-06 13:38:41.229+00:00 failed execution of retention policy       topics="waku archive" tid=1 file=archive.nim:257 error="decreaseDatabaseSize failed: decreaseDatabaseSize inside loop failed to remove oldest partition: error in ALTER TABLE messages DETACH PARTITION messages_1724454000_1724457600 CONCURRENTLY;: error in performWriteQuery: error in asyncpool query: error in dbConnQuery calling waitQueryToFinish: error in query: ERROR:  partition \"messages_1724454000_1724457600\" already pending detach in part"

Impact

The database runs out of space and then the store nodes keep restarting

nwaku version/commit hash

v0.36.0