There is something of a chicken-and-egg situation here. First and foremost, Wazuh collects logs in many ways. Check out this link: https://documentation.wazuh.com/current/user-manual/capabilities/log-data-collection/index.html. We can also collect and analyze remote events, for example, through syslog: https://documentation.wazuh.com/current/user-manual/capabilities/log-data-collection/syslog.html.
In any case, for a malicious process to make changes to the transport files, it is necessary to have bypassed the basic security provided by an environment with credentials. In other words, to gain access to modify files in the specified location, we must first find several potential events to be intercepted, such as a user login, a login rejection, etc.
However, it is possible to monitor some files or directories with FIM. But unless we include additional data like who-data (https://documentation.wazuh.com/current/user-manual/capabilities/file-integrity/advanced-settings.html#who-data-monitoring), it will not be possible to reach a conclusion.
The nature of your question is perfectly understandable, but for practical purposes, we must rely on other layers of security before checking for tampering in a logfile. Because if that user can modify the file, they could easily gain access to vital elements of Wazuh.