Slow SQL Instance Queries with Large Jobs Table

[MCterra10]

Describe the bug
Queries for details on server instances take a significant (>10 seconds) amount of time to complete.
Relevant slow queries log from MariaDB in logs section
Issue seems to be related to the ordering of the JOIN modifiers. I've had a similar issue when working with nfreader's statbus. The largest tables should be joined last to minimize processing, as joining large tables first expands the internal query table massively, creating processing delays for later joins, even if everything is eventually condensed for return.

To Reproduce
Steps to reproduce the behavior:

1. Create a large amount of rows in the jobs table
2. Connect to the control panel
3. Notice that the data takes a very long time to retrieve

Expected behavior
The queries take a minimal amount of time to execute and the control panel is snappy when loading instance details. Workaround is manually clearing out the jobs table, but this is only a temp fix.

Logs

# Time: 211208  8:51:45
# User@Host: tgs[tgs] @ localhost [127.0.0.1]
# Thread_id: 45805  Schema: tgs4  QC_hit: No
# Query_time: 10.170102  Lock_time: 0.000658  Rows_sent: 0  Rows_examined: 4510242
# Rows_affected: 0  Bytes_sent: 2296
SET timestamp=1638921105;
SELECT `j`.`Id`, `j`.`CancelRight`, `j`.`CancelRightsType`, `j`.`Cancelled`, `j`.`CancelledById`, `j`.`Description`, `j`.`ErrorCode`, `j`.`ExceptionDetails`, `j`.`InstanceId`, `j`.`StartedAt`, `j`.`StartedById`, `j`.`StoppedAt`, `u`.`Id`, `u`.`CanonicalName`, `u`.`CreatedAt`, `u`.`CreatedById`, `u`.`Enabled`, `u`.`GroupId`, `u`.`LastPasswordUpdate`, `u`.`Name`, `u`.`PasswordHash`, `u`.`SystemIdentifier`, `u0`.`Id`, `u0`.`CanonicalName`, `u0`.`CreatedAt`, `u0`.`CreatedById`, `u0`.`Enabled`, `u0`.`GroupId`, `u0`.`LastPasswordUpdate`, `u0`.`Name`, `u0`.`PasswordHash`, `u0`.`SystemIdentifier`, `i0`.`Id`, `i0`.`AutoUpdateInterval`, `i0`.`ChatBotLimit`, `i0`.`ConfigurationType`, `i0`.`Name`, `i0`.`Online`, `i0`.`Path`, `i0`.`SwarmIdentifer`
FROM `Instances` AS `i`
INNER JOIN `Jobs` AS `j` ON `i`.`Id` = `j`.`InstanceId`
INNER JOIN `Users` AS `u` ON `j`.`StartedById` = `u`.`Id`
LEFT JOIN `Users` AS `u0` ON `u`.`CreatedById` = `u0`.`Id`
INNER JOIN `Instances` AS `i0` ON `j`.`InstanceId` = `i0`.`Id`
WHERE `i`.`SwarmIdentifer` IS NULL AND (`j`.`StoppedAt` IS NULL AND (`j`.`Description` LIKE 'Move instance ID %'));

Server State: (please complete the following information):

• OS: Windows Server 2019 - v1809
• Version: TGS 4 v4.15.7
• Database Type/Version: MariaDB 10.6
• BYOND Version Used: N/A
• git Repository Used: austation/austation
• Origin Commit hash Used: N/A
• Active Test Merges: N/A
• Client Version: .NET control panel v3.1.0.0

Additional context
N/A

[Cyberboss]

Easiest solution is to implement job codes to differentiate job types and then index those