Sort tied categorical histogram frequencies alphabetically

[kyraman]

Description of changes:

Add lexicographic (alphabetical) secondary sort as a tiebreaker when categorical values share the same frequency count in the Histogram analyzer.

Before this change, if two categories had the same count, their ordering was non-deterministic (dependent on Spark's internal partition layout). For example, given Iris species data where "Iris-versicolor" and "Iris-virginica" both appear 50 times, the result could return either order across different runs.

After this change, tied values are sorted alphabetically, so the output is always:

Iris-setosa       (51)
Iris-versicolor   (50)   alphabetically before Iris-virginica
Iris-virginica    (50)

This matches the convention used by tools like pandas value_counts() and R's table(), which all use lexicographic ordering as a tiebreaker for equal counts.

Note: This change only affects the categorical Histogram analyzer. The numeric HistogramBinned analyzer, which uses equal-width bin edges, is unaffected.

By submitting this pull request, I confirm that my contribution is made under the terms of the Apache 2.0 license.