Optimize binary search methods

[IvoDD]

Reference Issues/PRs

Optimizations on top of #3091
Used in #3062

What does this implement or fix?

Some micro optimizations on binary search methods:

• Don't keep TypedBlockData in ColumnDataIterator. Instead only keep block_data_ and block_size_
• Don't recalculate block pointer and size when we already know them during gallop

Any other comments?

Benchmarks for all search and iteration methods:

Benchmark	Before (ns)	After (ns)	Delta
iterate_irregular_blocks_1 (one row per block)	478,496	311,163	−35.0%
iterate_with_iterator (100 rows)	798	719	−9.9%
exponential_lb_single_block (in first 100)	356	323	−9.2%
exponential_lb_single_block (full gallop)	458	424	−7.4%
exponential_lb_regular (in first 100)	364	339	−6.7%
exponential_lb_irregular_1000 (in first 100)	360	335	−6.7%
exponential_lb_irregular_1000 (full gallop)	496	476	−3.9%
exponential_lb_regular (full gallop)	504	489	−2.9%
exponential_lb_irregular_1 (in first 100)	464	455	−2.0%
exponential_lb_irregular_1 (full gallop)	687	679	−1.3%
lower_bound_single_block	411	394	−4.1%
lower_bound_irregular_1000	444	431	−3.0%
lower_bound_irregular_1	595	579	−2.8%
lower_bound_regular_blocks	443	436	−1.4%
iterate_single_block	27,305	27,247	−0.2%
iterate_regular_blocks	29,051	28,734	−1.1%
iterate_irregular_blocks_1000	28,136	27,893	−0.9%
iterate_with_scalar_at (100 rows)	182,183,122	182,088,026	−0.1%

Checklist

Checklist for code changes...

• Have you updated the relevant docstrings, documentation and copyright notice?
• Is this contribution tested against all ArcticDB's features?
• Do all exceptions introduced raise appropriate error messages?
• Are API changes highlighted in the PR description?
• Is the PR labelled as enhancement or bug so it appears in autogenerated release notes?

[claude]

ArcticDB Code Review Summary

No items requiring attention. The optimization is correct, well-scoped, and the benchmark deltas in the PR description validate it.

Verified:

• gallop_bracket first-block lambdas are safe: prev_block/cur_block remain first_block_idx throughout the first-block probing phase, so the optimized variants do not need to track the block field.
• The raw-pointer block_begin_ replacing std::optional<TypedBlockData<TDT>> is consistently propagated (copy constructor, dereference, end-sentinel). All callers in column_algorithms.hpp and test_column.cpp updated to current_block_data() == nullptr.
• New load_current_block computes block->logical_size() / sizeof(RawType), equivalent to the previous TypedBlockData::row_count() for Dim0 (which is static_assert-enforced by the search code paths).

[vasil-pashov]

Do these two extra assignments that are omitted really make a difference?

[IvoDD]

Most probably they do not.

Most of the benefit is from reusing the already calculated first_block_row_count and first_block_data in make_iter_in_first_block.

It made sense to also add a first block variant of record_probe as well, to make the invariant clearer