Simple templated classes for Metric Space representation.

The main class is template <class T, size_t SIZE> class MetricSpace::TMetricSpace.

It has two parameters:

1. used numeric type
2. space dimension

and 2 nested classes:

1. MetricSpace::TMetricSpace<>::Point - a point.
2. MetricSpace::TMetricSpace<>::Metric - a metric.

and four static functions, each returns one of predefined metrics:

1. MetricSpace::TMetricSpace<>::Euclidian() - returns Euclidean distance.
2. MetricSpace::TMetricSpace<>::SquaredEuclidian() - returns squared Euclidean distance.
3. MetricSpace::TMetricSpace<>::Manhattan() - returns Manhattan or taxicab.
4. MetricSpace::TMetricSpace<>::Chebyshev() - returns Chebyshev distance.

The main program (main.cpp) contains a "custom" metric "Weighted", parameterized by vector of "weights" of dimensions. The distance is calculated as a weighted sum of absolute dimensions differences.

The class MetricSpace::TMetricSpace was used to study cluster analysis.