Ordinal patterns: better documentation and add tests again

[kahaaga]

A suggestion to fix #242.

• Encodings are part of the public API now, so re-introduce tests.
• Add an isperm keyword to encode(::OrdinalPatternEncoding), so that the user can encode permutations directly too. If this option is unavailable, then the tests will fail, because the code calls sortperm on something which is already a permutation that would sort the input data. This is essentially sortperm(sortperm(x)), which destroys the one-to-one encoding-decoding. encode(::OrdinalPatternEncoding, x) behaves as before, but encode(::OrdinalPatternEncoding, x; isperm = true) skips the sortperm call.
• For OrdinalPatternEncoding, we test encoding/decoding against the example in table 1 in Berger et al. (2019).
• Updated the docstring for OrdinalPatternEncoding.

[codecov[bot]]

Codecov Report

Merging #243 (b49319f) into main (93362bf) will increase coverage by 0.28%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main     #243      +/-   ##
==========================================
+ Coverage   85.63%   85.91%   +0.28%     
==========================================
  Files          49       49              
  Lines        1211     1214       +3     
==========================================
+ Hits         1037     1043       +6     
+ Misses        174      171       -3     

Impacted Files	Coverage Δ
...c/probabilities_estimators/symbolic_permutation.jl	94.00% <ø> (ø)
src/encoding_implementations/ordinal_pattern.jl	86.79% <100.00%> (+6.79%)	:arrow_up:

:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more

[kahaaga]

@Datseris Ok, your comments are addressed.