[core] Introduce file index read/write framework.

[leaves12138]

• API:

CREATE TABLE <PAIMON_TABLE> (<COLUMN> <COLUMN_TYPE> , ...) WITH
(
"file-index.bloom-filter.columns" = "c1,c2",
"file-index.bloom-filter.c1.items" = "200",
"file-index.bloom-filter.c1.fpp" = "0.1",
"file-index.in-manifest-threshold" = "500 B" --default 500 B
)

• GOAL

This pull request, introduce the read and write framework to paimon. If the file index bytes size is smaller than file.index.size-in-meta, WRITE PART will put them in DataFileMeta. Otherwise, put them in index file. see #3068.

• Configurations

1. file-index.<index-Type>.columns: column names those should be indexed.
2. file-index.in-manifest-threshold: define the bytes size to store in DataFileMeta
3. file-index.read.enabled: true or false. Default true. Set this to false, read stage will skip file index check.
4. file-index.<index-Type>.<columnName>.<opKey>: set the property for specified index column.

[JingsongLi]

+1