Support BAI index aka BAM index index

[cmdcolin]

For some BAM files like nanopore and pacbio, the index file is regularly 100MB or bigger

There is the concept of a BAI index created here
https://github.com/hammerlab/bai-indexer/

This implementation doesn't include the chromosome names in the BAI index, so it is actually not really usable for us IMO. I don't think it's a good thing to rely on chromosomes being connected to the index in an array e.g. chr1 is position 0 in array

So we probably want our own BAI index, and also we could create a BAI index on the e.g. first download BAI, then index it and store it in localstorage or similar, then subsequent refreshes use the BAI index. This is good because chrome decides not to cache large BAI files in it's file cache. We can also have a on-disk representation of the BAI index so that this could be used instead of the localstorage also

[cmdcolin]

I had also discussed on desktop the entire index files could be cached (samtools for example does this) but having BAI index cache would basically be sufficient to cut down download times and would help both web and desktop

[cmdcolin]

Moved to bam-js GMOD/bam-js#52