more efficient building of sparse stoichiometries

[exaexa]

Description

This is for JSON and Standard models. MATModel can't be optimized, CoreModel contains the stoichiometry as is, and SBMLModel builds the stoichiometry in SBML.jl (I will optimize it there).

Most notably, this vastly reduces load time of large StdModels, roughly to O(json_parsing_time).

Checklist

• All code is formatted
• Tests work
• All new functions have proper docstrings (there are none)
• Documentation builds correctly (no changes)

[cylon-x]

🟢  Commit 73774a7 is formatted properly.

[codecov]

Codecov Report

Merging #443 (73774a7) into master (2df9398) will increase coverage by 0.28%.
The diff coverage is 95.65%.

@@            Coverage Diff             @@
##           master     #443      +/-   ##
==========================================
+ Coverage   89.47%   89.76%   +0.28%     
==========================================
  Files          59       59              
  Lines        1435     1465      +30     
==========================================
+ Hits         1284     1315      +31     
+ Misses        151      150       -1     

Impacted Files	Coverage Δ
src/base/utils/Reaction.jl	97.36% <83.33%> (-2.64%)	⬇️
src/base/types/SBMLModel.jl	85.71% <85.71%> (-0.34%)	⬇️
src/base/types/JSONModel.jl	100.00% <100.00%> (+1.31%)	⬆️
src/base/types/StandardModel.jl	100.00% <100.00%> (+1.72%)	⬆️
src/reconstruction/community.jl	91.72% <100.00%> (+0.62%)	⬆️

---

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 f20e41f...73774a7. Read the comment docs.

[stelmo]

:]