sourmash signature describe misses dayhoff, hp molecule types

[bluegenes]

Looks like signature describe isn't handling dayhoff, hp encoding properly. In the example below, these dayhoff & hp sigs are described as DNA sigs.

protein signature example, with protein, dayhoff, hp molecule encodings: test.sig:

[{"class":"sourmash_signature","email":"","hash_function":"0.murmur64","filename":"test.prot","license":"CC0","signatures":[{"num":0,"ksize":11,"seed":42,"max_hash":922337203685477632,"mins":                     [8654912015719540,36111891283354331,37308772364955832,87386056322424681,172394428206145264,175324458129672500,192458836906930061,203682007118818914,228487149606072746,249673742953190691,252759730738350599,       269851384402800409,276879657160814740,350146213620382349,355128279925181821,362142414738517052,372280419647394145,378568520451171994,400417693755427902,411540066123110581,447785525058093308,                      508503262931135755,558153676748293637,603380202800315060,608284974085382699,672722089804872372,703845239558431174,717802877765604512,725736535145543392,773956917118790885,802909586863648861,                      874660063252901754,877868933880276449,891726720659491306],"md5sum":"57ae47c24a08a24b630dae80f0c6e256","abundances":[1,2,2,1,2,1,1,1,1,1,1,1,1,2,1,1,1,1,1,1,1,1,1,2,1,1,1,1,2,1,1,1,1,1],"molecule":"protein"},     {"num":0,"ksize":11,"seed":42,"max_hash":922337203685477632,"mins":[394785509353664721,461721674775207677,709435529205169740,851683204813363936],"md5sum":"407a774f7a2e13c650ff67255d2056f8","abundances":[2,4,     1,10],"molecule":"dayhoff"},{"num":0,"ksize":11,"seed":42,"max_hash":922337203685477632,"mins":[48718379107742671,311269990247623707],"md5sum":"72560192a7f4c23f0b4d71b3eaaa44db","abundances":[112,89], "molecule":"hp"}],"version":0.4}]

Running sourmash signature describe on this:

$ sourmash signature describe test.sig

== This is sourmash version 3.2.2. ==
== Please cite Brown and Irber (2016), doi:10.21105/joss.00027. ==

loaded 3 signatures total.
---
signature filename: test.sig
signature: test.prot
source file: test.prot
md5: 57ae47c24a08a24b630dae80f0c6e256
k=11 molecule=protein num=0 scaled=20 seed=42 track_abundance=1
size: 34
signature license: CC0

---
signature filename: test.sig
signature: test.prot
source file: test.prot
md5: 407a774f7a2e13c650ff67255d2056f8
k=11 molecule=DNA num=0 scaled=20 seed=42 track_abundance=1
size: 4
signature license: CC0

---
signature filename: test.sig
signature: test.prot
source file: test.prot
md5: 72560192a7f4c23f0b4d71b3eaaa44db
k=11 molecule=DNA num=0 scaled=20 seed=42 track_abundance=1
size: 2
signature license: CC0

[olgabot]

Uh oh! I'm betting this an issue here: https://github.com/dib-lab/sourmash/blob/master/sourmash/sig/__main__.py#L101 which I think should be using get_moltype instead: https://github.com/dib-lab/sourmash/blob/1c617da0385b451d26a8bae4e0b33f26e32a2fe5/sourmash/sourmash_args.py