fix array variables, brusselator test, periodic wraparound

[xtalax]

fixes #43
fixes #47
Adds ODEFunctionExpr constructor and generate_code method.

[codecov]

Codecov Report

Merging #45 (d8881b6) into master (1374af2) will decrease coverage by 34.42%.
The diff coverage is 61.02%.

@@             Coverage Diff             @@
##           master      #45       +/-   ##
===========================================
- Coverage   92.90%   58.47%   -34.43%     
===========================================
  Files          11       12        +1     
  Lines         761      814       +53     
===========================================
- Hits          707      476      -231     
- Misses         54      338      +284     

Impacted Files	Coverage Δ
src/discretization/differential_discretizer.jl	38.75% <22.22%> (-58.85%)	⬇️
src/bcs/generate_bc_eqs.jl	35.10% <45.16%> (-44.67%)	⬇️
src/bcs/boundary_types.jl	74.22% <57.14%> (-24.63%)	⬇️
src/interiormap.jl	74.76% <62.50%> (-16.50%)	⬇️
...discretization/generate_finite_difference_rules.jl	55.42% <64.28%> (-39.76%)	⬇️
src/MOL_utils.jl	73.01% <88.23%> (-16.27%)	⬇️
src/discretization/MOL_discretization.jl	64.07% <91.66%> (-35.93%)	⬇️
src/discretization/discretize_vars.jl	67.18% <100.00%> (-21.88%)	⬇️
src/periodic_map.jl	100.00% <100.00%> (ø)
src/discretization/fornberg.jl	0.00% <0.00%> (-100.00%)	⬇️
... and 8 more

📣 Codecov can now indicate which changes are the most critical in Pull Requests. Learn more

[ChrisRackauckas]

How do you know this is producing the right result? The test doesn't seem specific enough.

[xtalax]

I don't, I figured that since the only thing it changes is variable names if the test even runs it's pretty much guaranteed to work. I can try it though

[ChrisRackauckas]

Array variables are scary. I'd definitely double check that assumption 😅