Allow adding a unique suffix to segment name during batch ingestion

[KKcorps]

Currently, if a user creates segments from the batch data that

• contains either no time column or
• contains same time for all the records (e.g. Current data in yyyy-MM-dd format)

the generated segments have same names. This creates issue because the segments end up replacing each other even when the push type is set to APPEND

The solution is to either add a current timestamp column in the table or provide a config in SegmentNameGenerationSpec to append currentTimestamp, uuid or something else at the end of the segment name

[KKcorps]

InputFileSegmentNameGenerator handles only one scenario whereby it can create a unique segment per file. But if file names are same in subsequent runs of the Generation job, it can still create issues.

[xiangfu0]

This makes sense, I think we can add a config to append creation time or random sequence_id to make the segment name unique.