You have different chromosome names in your reference fasta file than in your .vcf file.
The reference fasta contains genbank ids (AP014957.1) while the VCF contains chromosome numbers "1", "2", ...
Edit: You could rename the sequences in your reference file (AP014957.1 = chromosome 1):
>AP014957.1
to look like:
>1
....
>2
You can either do this manually, it doesn't really pay off to automate this for 12 chromosomes or try to download a different reference genome. In particular, check the VCF file on which reference it is based. The info should be in the VCF, too.
Likely, you must do the initial alignment from scratch if you used a different reference.
Before you go ahead, make sure to use the identical reference in all steps. If you want to use base quality recalibration, you need to use the same reference as in the know-sites.vcf in all other steps.
I would personally not bother with renaming VCF or BAM entries if avoidable; chances are that this messes things up. On the other hand, if you intend to run variant annotation after that, you also need to check that the reference of the annotation file used in any subsequent process needs to be compatible and based on the identical assembly, too.