Show documents that might be around after N1QL emptying fails

[torcolvin]

In some weekly-jenkins runs we see errors like:

2023-12-14T21:04:28.868Z TEST: b:sg_int_1_1702585440608119514 waiting for empty bucket indexes sg_int_1_1702585440608119514._default._default
2023-12-14T21:09:16.752Z TEST: b:sg_int_1_1702585440608119514 waitForPrimaryIndexEmpty returned an error: RetryLoop for Wait for index to be empty giving up after 61 attempts
2023-12-14T21:09:16.752Z TEST: b:sg_int_1_1702585440608119514 Couldn't ready bucket, got error: RetryLoop for Wait for index to be empty giving up after 61 attempts - Retrying
2023-12-14T21:09:16.752Z [WRN] b:sg_int_1_1702585440608119514 RetryLoop for Wait for index to be empty giving up after 61 attempts -- base.RetryLoop() at util.go:460
2023-12-14T21:09:18.753Z TEST: b:sg_int_1_1702585440608119514 Running bucket through readier function
2023-12-14T21:09:18.753Z TEST: b:sg_int_1_1702585440608119514 emptying bucket via N1QL, readying views and indexes
2023-12-14T21:09:18.765Z TEST: b:sg_int_1_1702585440608119514 Bucket not empty (2 items), emptying bucket via N1QL
2023-12-14T21:09:18.783Z TEST: b:sg_int_1_1702585440608119514 Finished emptying all docs index ... Total docs purged: 0
2023-12-14T21:09:18.792Z TEST: b:sg_int_1_1702585440608119514 waiting for empty bucket indexes sg_int_1_1702585440608119514._default._default

You can see here that it has gone on for 61 times and is fact growing documents. https://jenkins.sgwdev.com/job/weekly-matrix/BACKING_STORE=enterprise-7.1.6,QUERY_PROVIDER=gsi-default-collection,SYNC_STORE=inline,label=sgw-amazon-linux2023/100/

In this case, the test that seems to be leaking is: TestDbConfigPersistentSGVersions and I have to imagine that is config files that are leaking but I don't see how this test is potentially leaking goroutines. Note, I'm modifying this test in #6622 to make it work with rosmar.

Tested by removing N1QL delete in bucketReadier and making sure we get a better log message.

Pre-review checklist

• Removed debug logging (fmt.Print, log.Print, ...)
• Logging sensitive data? Make sure it's tagged (e.g. base.UD(docID), base.MD(dbName))
• Updated relevant information in the API specifications (such as endpoint descriptions, schemas, ...) in docs/api