Duplicate download detection seems not to be working

[MattBlissett]

We have code that is supposed to prevent a user from submitting identical download requests in a short interval:

occurrence/occurrence-download-service/src/main/java/org/gbif/occurrence/download/resource/DownloadResource.java

Lines 374 to 385 in 1911143

	// Check for recent user downloads (of this same user) with the same predicate
	PagingResponse<Download> userDownloads =
	occurrenceDownloadService.listByUser(
	userAuthenticated.getName(),
	new PagingRequest(0, 50),
	EnumSet.of(PREPARING, RUNNING, SUCCEEDED),
	LocalDateTime.now().minus(4, ChronoUnit.HOURS),
	false);
	String existingUserDownload = matchExistingDownload(userDownloads, predicateDownloadRequest);
	if (existingUserDownload != null) {
	return existingUserDownload;
	}

However, there are duplicate downloads — hash is the MD5sum of the filter:

  status   │           key           │     format     │ size_gb │            created            │               md5                │              
───────────┼─────────────────────────┼────────────────┼─────────┼───────────────────────────────┼──────────────────────────────────┼
 SUCCEEDED │ 0264549-230224095556074 │ SIMPLE_CSV     │     212 │ 2023-05-28 13:58:03.559294+00 │ 4d98ea545c5b5cc7fa1fcfbee587dbd8 │
 SUCCEEDED │ 0264553-230224095556074 │ SIMPLE_CSV     │     212 │ 2023-05-28 14:02:22.877401+00 │ 4d98ea545c5b5cc7fa1fcfbee587dbd8 │
 SUCCEEDED │ 0264554-230224095556074 │ SIMPLE_CSV     │     212 │ 2023-05-28 14:02:30.244787+00 │ 4d98ea545c5b5cc7fa1fcfbee587dbd8 │
 SUCCEEDED │ 0264567-230224095556074 │ SIMPLE_CSV     │     212 │ 2023-05-28 14:33:13.458542+00 │ 4d98ea545c5b5cc7fa1fcfbee587dbd8 │
 SUCCEEDED │ 0264573-230224095556074 │ SIMPLE_CSV     │     212 │ 2023-05-28 14:46:41.546757+00 │ 4d98ea545c5b5cc7fa1fcfbee587dbd8 │
 SUCCEEDED │ 0264574-230224095556074 │ SIMPLE_CSV     │     212 │ 2023-05-28 14:46:49.502762+00 │ 4d98ea545c5b5cc7fa1fcfbee587dbd8 │

 SUCCEEDED │ 0159495-230224095556074 │ SIMPLE_CSV     │     264 │ 2023-04-11 06:59:10.942603+00 │ 7f8d83b63d0c620de6f366741724597a │ 
 SUCCEEDED │ 0159570-230224095556074 │ SIMPLE_CSV     │     264 │ 2023-04-11 07:28:32.897055+00 │ 7f8d83b63d0c620de6f366741724597a │ 
 SUCCEEDED │ 0159571-230224095556074 │ SIMPLE_CSV     │     264 │ 2023-04-11 07:28:41.082502+00 │ 7f8d83b63d0c620de6f366741724597a │ 

 SUCCEEDED │ 0260165-230224095556074 │ DWCA           │     208 │ 2023-05-24 20:31:03.228036+00 │ be9885a3a2ecc9fdf7eecd22900f4986 │ 
 SUCCEEDED │ 0260166-230224095556074 │ DWCA           │     208 │ 2023-05-24 20:31:10.732704+00 │ be9885a3a2ecc9fdf7eecd22900f4986 │ 

This wastes a lot of space, we should check why the code isn't working, and should probably extend the time for duplicates from 4 to 24 hours.