Not sure why you are using SILVA for ITS reads. SILVA only contains sequence data for SSU rRNA genes (i.e. 16S & 18S). So, I would not trust any classifications of ITS data resulting from a search against the SILVA db.
Could you provide more details on what primer sets you used? Did you merge paired-end reads, or just use the forward read? For example see this post of helpful questions:
Perhaps provide a list of commands, or simply the QZV file of the taxonomy barplot so we can look at the provenance information. Feel free to DM me if you do not want to share publicly.
Finally, if you search the forum you'll come across quite a few threads about Fungal ITS analysis. Here is one to get you started: