ENH: state space: compute smoothed state autocovariance matrices for arbitrary lags

[ChadFulton]

• tests added / passed.
• code/documentation is well formatted.
• properly formatted commit message.

We already compute the lag=1 smoothed autocovariance matrix for the state vector (i.e. Cov(t, t-1)) in the Cython recursions (because it is required for the EM algorithm applied to the transition equation, and so it's best if it's pretty fast).

This PR adds a new smoothed_state_autocovariance method to the SmootherResults class to allow computing autocovariances with any lag = 0, +-1, +-2, .... One application of this is computing the "news" for nowcasting models, described in Banbura and Modugo (2014). Not in Cython since as far as I know, this is just for postestimation results, and won't be in an inner loop.

[codecov]

Codecov Report

Merging #6579 into master will increase coverage by 0.02%.
The diff coverage is 93.14%.

@@            Coverage Diff             @@
##           master    #6579      +/-   ##
==========================================
+ Coverage   85.28%   85.31%   +0.02%     
==========================================
  Files         646      646              
  Lines      103651   103922     +271     
  Branches    11270    11311      +41     
==========================================
+ Hits        88401    88657     +256     
- Misses      12797    12806       +9     
- Partials     2453     2459       +6

Impacted Files	Coverage Δ
statsmodels/tsa/statespace/tests/test_smoothing.py	98.83% <100%> (+0.5%)	⬆️
...els/tsa/statespace/tests/test_impulse_responses.py	100% <100%> (ø)	⬆️
statsmodels/tsa/statespace/kalman_smoother.py	86.09% <76.54%> (-3.1%)	⬇️
statsmodels/tsa/statespace/representation.py	88.13% <0%> (+0.56%)	⬆️

---

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 3c74781...37272a4. Read the comment docs.

[ChadFulton]

Squashed.

[coveralls]

Coverage increased (+0.03%) to 87.661% when pulling 0b85a4d on ChadFulton:ss-acov-lgt1 into c0e585d on statsmodels:master.

[coveralls]

Coverage increased (+0.01%) to 87.674% when pulling 37272a4 on ChadFulton:ss-acov-lgt1 into 3c74781 on statsmodels:master.

[ChadFulton]

Also, this is still a semi-private method, since this PR only puts a new method in SmootherResults and doesn't make it available as a method in MLEResults, which I think is good to start with.