Refactor Poly class and its inheritance

[mhostetter]

This PR implements #176. As a consequence of removing DensePoly, BinaryPoly, and SparsePoly and consolidating them into Poly, binary array arithmetic/creation is much faster. Below are some examples.

Polynomial arithmetic over GF(2)

In [1]: import galois

In [2]: f = galois.Poly.Random(20, seed=1); f
Out[2]: Poly(x^20 + x^19 + x^16 + x^15 + x^13 + x^11 + x^9 + x^8 + x^7 + x^5 + x^4 + x^2 + 1, GF(2))

In [3]: g = galois.Poly.Random(10, seed=1); g
Out[3]: Poly(x^10 + x^9 + x^6 + x^5 + x^3 + x, GF(2))

In [4]: f * g
Out[4]: Poly(x^30 + x^28 + x^22 + x^20 + x^19 + x^13 + x^11 + x^10 + x^9 + x^8 + x^5 + x, GF(2))

# Before
In [5]: %timeit f * g
196 µs ± 11.4 µs per loop (mean ± std. dev. of 7 runs, 1,000 loops each)

# After
In [5]: %timeit f * g
2.66 µs ± 13 ns per loop (mean ± std. dev. of 7 runs, 100,000 loops each)

Polynomial search routines in GF(2)

# Before
In [4]: %timeit galois.primitive_polys(2, 12)
10.1 s ± 123 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)

# After
In [2]: %timeit galois.primitive_polys(2, 12)
783 ms ± 1.53 ms per loop (mean ± std. dev. of 7 runs, 1 loop each)

[codecov]

Codecov Report

Merging #320 (c8a753d) into master (a454532) will decrease coverage by 0.61%.
The diff coverage is 86.08%.

@@            Coverage Diff             @@
##           master     #320      +/-   ##
==========================================
- Coverage   93.73%   93.11%   -0.62%     
==========================================
  Files          30       30              
  Lines        5156     5128      -28     
==========================================
- Hits         4833     4775      -58     
- Misses        323      353      +30     

Impacted Files	Coverage Δ
galois/_fields/_main.py	90.67% <84.47%> (-2.18%)	⬇️
galois/_poly_conversion.py	97.22% <95.74%> (-0.96%)	⬇️
galois/_fields/_functions.py	88.27% <100.00%> (ø)
galois/_fields/_poly_functions.py	100.00% <100.00%> (ø)
galois/_fields/_ufuncs.py	88.55% <0.00%> (-0.61%)	⬇️

---

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 a454532...c8a753d. Read the comment docs.