bcftools norm -m +snps doesn't combine INFO fields when combining identical SNPs

Start with the bcftools norm -m +snps handling described in the issue and reproduce the behavior using tiny.vcf. The change is done when identical SNP records combine their INFO values as requested, including summed AC and AF, while the different-allele example continues to produce comma-separated values.

Below is a tiny vcf file containing two data records, each describing the same SNP. When I run bcftools norm -m +snps on it, it properly outputs only one data record, but it doesn't combine the INFO fields-- it just takes the INFO field from the second data record.

$ cat tiny.vcf
##fileformat=VCFv4.1
##FILTER=<ID=PASS,Description="All filters passed">
##contig=<ID=1,URL=ftp://somewhere.org/assembly.fa>
##INFO=<ID=AF,Number=A,Type=Float,Description="Allele Frequency">
##INFO=<ID=AC,Number=A,Type=Integer,Description="Allele Count">
##INFO=<ID=AN,Number=1,Type=Integer,Description="Total number of alleles in data sources">
##INFO=<ID=END,Number=.,Type=Integer,Description="End position">
##INFO=<ID=DS,Number=A,Type=String,Description="Data Sources containing allele">
##INFO=<ID=OLD_MULTIALLELIC,Number=1,Type=String,Description="Original chr:pos:ref:alt encoding">
##INFO=<ID=OLD_VARIANT,Number=.,Type=String,Description="Original chr:pos:ref:alt encoding">
#CHROM  POS     ID      REF     ALT     QUAL    FILTER  INFO
1       10440   .       C       A       .       .       AF=0.0002305;AC=6;AN=26028
1       10440   .       C       A       .       .       AF=0.0001153;AC=3;AN=26028

$ bcftools norm -m+snps tiny.vcf  #header info omitted
1       10440   .       C       A       .       .       AF=0.0001153;AC=3;AN=26028

bcftools norm -m +snps does combine INFO fields when combining two different SNP alleles at the same position--it combines the AC and AF values by separating them with commas (see below).

$ cat tiny2.vcf #header info same as above; omitted
1       10440   .       C       A       .       .       AF=0.0002305;AC=6;AN=26028
1       10440   .       C       T       .       .       AF=0.0001153;AC=3;AN=26028

$  bcftools norm -m+snps tiny2.vcf
1       10440   .       C       A,T     .       .       AF=0.0002305,0.0001153;AC=6,3;AN=26028

I'd like it to also do the right thing when the two alleles are the same--I'd like it to sum the AC and AF values, producing the following for tiny.vcf:

1       10440   .       C       A       .       .       AF=0.0003458;AC=9;AN=26028

I am using this version of bcftools:

$ bcftools --version
bcftools 1.2-190-g71921ca
Using htslib 1.2.1-226-g1e5c377