refactor: migrate TSV input format to new framework.

[youngsofun]

I hereby agree to the terms of the CLA available at: https://docs.databend.com/dev/policies/cla/

Summary

1. no longer tries to read big TSV in parallel to make it simple and always able to get row id, but still can be deserialized in parallel after cutting into RowBatches.
2. simplify the handling of skip headers.
3. minimal reallocating of file data.

Tests

• [ ] Unit Test
• [x] Logic Test
• [ ] Benchmark Test
• [ ] No Test - Explain why

Type of change

• [ ] Bug Fix (non-breaking change which fixes an issue)
• [ ] New Feature (non-breaking change which adds functionality)
• [ ] Breaking Change (fix or feature that could cause existing functionality not to work as expected)
• [ ] Documentation Update
• [x] Refactoring
• [ ] Performance Improvement
• [ ] Other (please describe):

---

This change is 

[github-actions[bot]]

Docker Image for PR

• tag: pr-15506-2ea3cd2

note: this image tag is only available for internal use, please check the internal doc for more details.

[github-actions[bot]]

ClickBench Report

• hits: https://benchmark.databend.rs/clickbench/pr/15506/9090107735/hits.html
• internal: https://benchmark.databend.rs/clickbench/pr/15506/9090107735/internal.html
• load: https://benchmark.databend.rs/clickbench/pr/15506/9090107735/load.html
• tpch: https://benchmark.databend.rs/clickbench/pr/15506/9090107735/tpch.html

[youngsofun]

Q2 is loading a large TSV, slightly slower, I think it is negligible.