I would not use fastq-mcf to trim primer sequences. Primers will be present in their entirety, whereas adaptors are only partially present, if at all. The best example of highly necessary and high-quality adatptor trimming is when you're doing miRNA (21bp sequences with 35bp or 50bp reads), where partial adaptors are present on every read.
I would just align with something like bwamem against the primers, remove the aligned bits, and then re-align against the genome. (don't align to targetome! )