The non-UTF8 character check excludes valid UTF8 characters

[zbalkan]

Wazuh version	Component	Install type	Install method	Platform
4.7.5>	Wazuh agent	Agent	N/A	N/A

Summary

This is a continuation of issue wazuh/wazuh#23354, about the fix PR wazuh/wazuh#23543.

After the fix, the agent, unfortunately, does an ASCII character check instead of UTF-8 check. Therefore, when there are characters like Ü, ü, Õ, õ, Ö, ö, Ä, ä, Ş, ş, Ç, ç, etc., in the path of the files, the FIM ignores them as not valid UTF-8 while they are valid UTF-8 characters. These are valid UTF-8 characters and I believe the nlohmann library would accept these as well.