Add Sobol indices and polyharmonic splines

[FriesischScott]

This adds the following new features:

• Polyharmonic splines as a metamodel
• Computation of first order and total effect Sobol indices using Monte Carlo simulation
  • Quasi Monte Carlo methods return incorrect results for some reason
• Sobol sampling
• Halton sampling

[codecov-io]

Codecov Report

Merging #18 (da2eadc) into master (bea5015) will decrease coverage by 0.42%. The diff coverage is 98.93%.

@@             Coverage Diff             @@
##            master      #18      +/-   ##
===========================================
- Coverage   100.00%   99.57%   -0.43%     
===========================================
  Files           11       13       +2     
  Lines          140      233      +93     
===========================================
+ Hits           140      232      +92     
- Misses           0        1       +1     

Impacted Files	Coverage Δ
src/UncertaintyQuantification.jl	100.00% <ø> (ø)
src/inputs/copulas/gaussian.jl	100.00% <ø> (ø)
src/inputs/inputs.jl	100.00% <ø> (ø)
src/inputs/jointdistribution.jl	100.00% <ø> (ø)
src/inputs/parameter.jl	100.00% <ø> (ø)
src/inputs/randomvariable.jl	100.00% <ø> (ø)
src/reliability/probabilityoffailure.jl	100.00% <ø> (ø)
src/simulations/linesampling.jl	100.00% <ø> (ø)
src/models/polyharmonicspline.jl	96.55% <96.55%> (ø)
src/sensitivity/sobolindices.jl	100.00% <100.00%> (ø)
... and 3 more

---

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 bea5015...da2eadc. Read the comment docs.

[FriesischScott]

@AnderGray Apart from the issue with QMC (that I can't figure out) I think this is ready to merge for now. Do you want to have a look or should I just merge it as is?

[AnderGray]

It looks very nice!

We can take a look at the QMC at a future date.