Refactoring stumpy.motifs.match and improving its flexibility!

[NimaSarajpoor]

It seems that the functions stumpy.motifs.match(Q, T, ...) and stumpy.aamp_motifs.aamp_match(Q, T, ...) have the same structure, respectively, after line 460 and line 393.

So, I am thinking of refactoring as follows:

# in stumpy/core.py
# ADD NEW FUNCTION
def _match_finder(D, other_required_parameters):
     """
     Parameters
     ------------
     D is a 1D distance profile array
     other_required_parameters: max_distance,  max_matches, ...
      
     Returns
     --------
     out : numpy.ndarray
        The first column consists of distances of subsequences of `T` whose distances
        to `Q` are less than or equal to `max_distance`, sorted by distance (lowest to
        highest). The second column consists of the corresponding indices in `T`.
     """
     # lines 461-483 of the function `match` will be used here.

Then,

# in `stumpy.motifs`
def match(......, D=None):
     # add new parameter `D` so users can use pre-calculated distance profile
     if D is not None:
           # check if dimensions of `D` is correct and through a warning or error otherwise.
    else:
          # calculate D (see lines 422-460)
   
    return _match_finder(D, other_required_parameters)


# NOTE: we do the same thing for `stumpy.aamp_motifs`

The advantages:

• we refactored the functions
• one can use stumpy.mpdist._mpdist_vect to calculate mpdist-based distance profile and then pass it to _match_finder(D) to find mpdist-based matches! (they may use even other distances to get the distance profile, and then they can use _match_finder(D) to find the matches)

[seanlaw]

Sounds good. Please create an issue and maybe let's call the function _find_matches (in core.py)