Do add_many in Rust, use it in LCA _signatures

[luizirber]

Calling .add_hash() on a MinHash sketch is fine, but if you're calling it all the time it's better to pass a list of hashes and call .add_many() instead. Before this PR add_many just called add_hash for each hash it was passed, but now it will pass the full list to Rust (and that's way faster).

No changes for public APIs, and I changed the _signatures method in LCA to accumulate hashes for each sig first, and then set them all at once. This is way faster, but might use more intermediate memory (I'll evaluate this now).

Checklist

• Is it mergeable?
• make test Did it pass the tests?
• make coverage Is the new code covered?
• Did it change the command-line interface? Only additions are allowed
  without a major version increment. Changing file formats also requires a
  major version number increment.
• Was a spellchecker run on the source code and documentation after
  changes were made?

[codecov]

Codecov Report

Merging #826 into master will increase coverage by 0.03%.
The diff coverage is 100%.

@@            Coverage Diff             @@
##           master     #826      +/-   ##
==========================================
+ Coverage   79.17%   79.21%   +0.03%     
==========================================
  Files          45       45              
  Lines        6705     6707       +2     
  Branches      469      469              
==========================================
+ Hits         5309     5313       +4     
+ Misses       1096     1094       -2     
  Partials      300      300

Flag	Coverage Δ
#pytests	90.42% <100%> (+0.04%)	⬆️
#rusttests	48.71% <ø> (ø)	⬆️

Impacted Files	Coverage Δ
sourmash/lca/lca_utils.py	96.88% <100%> (+0.02%)	⬆️
sourmash/_minhash.py	97.81% <100%> (-0.01%)	⬇️
sourmash/utils.py	75.43% <0%> (+3.5%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 7e99e85...c6cbdf0. Read the comment docs.

[luizirber]

No changes for public APIs, and I changed the _signatures method in LCA to accumulate hashes for each sig first, and then set them all at once. This is way faster, but might use more intermediate memory (I'll evaluate this now).

As expected, it's using more memory. I tried used both a set and a list to accumulate hashes.

version	mem	time
original	1.5 GB	160s
set	3.8GB	80s
list	1.7GB	73s

So I kept the list version, since the memory increase is not so bad (and it's faster than the set).