Fix "extra dimension" in distribution.calc_expectations

[Mv77]

Fixes #1098

• Tests for new functionality/models or Tests to reproduce the bug-fix in code.
• Updated documentation of features that add new functionality.
• Update CHANGELOG.md with major/minor changes.

[Mv77]

I've fixed the function for now. I'm working on updating the many-many calls that took this issue as given.

[Mv77]

This is ready to merge. (Tests will pass, I just updated the changelog).

The one thing that becomes more ugly as a result of this PR is that calc_expectation(X) where X is a univariate distribution will return a size-1 array, not a scalar. This is the cost of making everything based around arrays and not having if-else statements figuring out the types of multiple possible inputs. It is a cost I'd gladly pay. That is why you will see, in places like check_conditions, things like calc_expectations(PermShkDstn)[0].

[codecov-commenter]

Codecov Report

Merging #1149 (4c7e780) into master (2155175) will decrease coverage by 0.01%.
The diff coverage is 96.05%.

@@            Coverage Diff             @@
##           master    #1149      +/-   ##
==========================================
- Coverage   74.07%   74.06%   -0.02%     
==========================================
  Files          70       70              
  Lines       10809    10807       -2     
==========================================
- Hits         8007     8004       -3     
- Misses       2802     2803       +1     

Impacted Files	Coverage Δ
HARK/ConsumptionSaving/ConsGenIncProcessModel.py	82.45% <ø> (+0.13%)	⬆️
HARK/ConsumptionSaving/ConsPortfolioModel.py	88.69% <ø> (-0.22%)	⬇️
...RK/ConsumptionSaving/tests/test_ConsMarkovModel.py	86.56% <ø> (ø)
...K/ConsumptionSaving/tests/test_modelcomparisons.py	98.46% <ø> (ø)
HARK/distribution.py	84.41% <90.32%> (-0.98%)	⬇️
HARK/ConsumptionSaving/ConsIndShockModel.py	85.86% <100.00%> (ø)
HARK/ConsumptionSaving/ConsLaborModel.py	82.90% <100.00%> (ø)
HARK/ConsumptionSaving/ConsMedModel.py	74.04% <100.00%> (ø)
HARK/ConsumptionSaving/ConsPrefShockModel.py	79.68% <100.00%> (ø)
HARK/ConsumptionSaving/ConsRiskyContribModel.py	79.42% <100.00%> (ø)
... and 5 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 704bf8e...4c7e780. Read the comment docs.

[alanlujan91]

Looks great!

[Mv77]

@llorracc @sbenthall I'm merging this in given that it is minor and passed tests and Alan's review.

Thanks @alanlujan91 !