[FEATURE] Enrich skipping index data structure

[dai-chen]

Is your feature request related to a problem?

In Flint 0.1, there are 3 skipping data structures: Partition, Min-Max, and ValueSet. Each of them serves specific purposes while also having its own set of advantages and limitations. For instance, the ValueSet can be incredibly useful but tends to consume a significant amount of space when handling columns with high cardinality.

What solution would you like?

Enrich the support for different skipping index type (reference paper <<Extensible Data Skipping>>):

1. BloomFilter: store a bloom filter structure [FEATURE] Add bloom filter skipping index type #206
2. Prefix: store the unique prefixes having b1 characters
3. Suffix: store the unique suffixes having b2 characters
4. GapList: store k gaps indicating ranges where there are no data points
5. GeoBox: store a set of x bounding boxes covering data points

As the range of supported data structures continues to expand, an important consideration emerges: the automated selection and evolution from one structure to another based on the accumulation of more data over time. This work is tracked in #118.