Additional Features for CubicHermiteInterp

[alanlujan91]

This PR adds new methods to CubicHermiteInterp.

Because CubicHermiteInterp uses scipy's CubicHermiteSpline, we can easily port additional methods that could be useful in HARK.

These were initially in #1011 but were moved to a standalone PR for further discussion.

These new methods are:

1. der_interp(self[, nu]) Construct a new piecewise polynomial representing the derivative.
2. antider_interp(self[, nu]) Construct a new piecewise polynomial representing the antiderivative.
3. integrate(self, a, b[, extrapolate]) Compute a definite integral over a piecewise polynomial.
4. roots(self[, discontinuity, extrapolate]) Find real roots of the the piecewise polynomial.
5. solve(self[, y, discontinuity, extrapolate]) Find real solutions of the the equation pp(x) == y.

See notebook for more details: https://github.com/alanlujan91/HARK/blob/CHI_new_feats/examples/Interpolation/CubicInterp.ipynb

[codecov-commenter]

Codecov Report

Merging #1020 (2f05061) into master (d59269d) will decrease coverage by 0.02%.
The diff coverage is 66.66%.

@@            Coverage Diff             @@
##           master    #1020      +/-   ##
==========================================
- Coverage   72.53%   72.51%   -0.03%     
==========================================
  Files          68       68              
  Lines       10298    10368      +70     
==========================================
+ Hits         7470     7518      +48     
- Misses       2828     2850      +22     

Impacted Files	Coverage Δ
HARK/interpolation.py	42.57% <65.78%> (+1.04%)	⬆️
HARK/ConsumptionSaving/ConsIndShockModel.py	85.86% <100.00%> (+0.01%)	⬆️
HARK/tests/test_interpolation.py	100.00% <100.00%> (ø)

---

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 d59269d...2f05061. Read the comment docs.

[sbenthall]

@alanlujan91 can you fix the merge conflict? (sorry for duplicating my comment on #1011 )

@mnwhite would you be able to review this?

[mnwhite]

I'm working on this now. GitHub flagged a bunch of things as "conflicts" that plainly were not. And it also got upset about notebook execution metadata, which is annoying. It's quite likely that I broke stuff in that commit.

[mnwhite]

This has one failing test, and I'm actually not sure whether it should pass. It looks like the test is for whether new cubic interpolator can accepts inputs of shape (N,1) rather than just (N,)-- flat 2D arrays. Is that intended?

[mnwhite]

The problem was some selfs that were removed. The init code does a "flat check", but its results weren't being used.