perf: optimize to_local_time in datafusion-functions

[andygrove]

This PR was created by an LLM as a draft PR. I will mark it as ready for review after human review.

Which issue does this PR close?

N/A — autonomous exploratory PR.

Rationale for this change

Replaced chrono DateTime/TimeDelta calendar arithmetic in the per-row to_local_time hot path with an exact integer offset shift on the raw i64 timestamp, removing the TimeDelta construction, the DateTime addition, and the convert-back per element.

What changes are included in this PR?

Replaced chrono DateTime/TimeDelta calendar arithmetic in the per-row to_local_time hot path with an exact integer offset shift on the raw i64 timestamp, removing the TimeDelta construction, the DateTime addition, and the convert-back per element.

Are these changes tested?

Correctness: unit tests + seeded differential fuzz (bit-identical Arrow output vs main).

Benchmark (criterion):

• to_local_time_no_nulls_100k: 8.891% faster (base 4139190ns -> cand 3771157ns)
• to_local_time_10pct_nulls_100k: 8.45% faster (base 3828852ns -> cand 3505310ns)

Full criterion output:

to_local_time_no_nulls_100k
                        time:   [3.7694 ms 3.7712 ms 3.7741 ms]
                        change: [−9.2033% −8.8914% −8.6695%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 2 outliers among 100 measurements (2.00%)
  2 (2.00%) high severe

to_local_time_10pct_nulls_100k
                        time:   [3.5037 ms 3.5053 ms 3.5080 ms]
                        change: [−8.5448% −8.4501% −8.3570%] (p = 0.00 < 0.05)
                        Performance has improved.
Found 1 outliers among 100 measurements (1.00%)
  1 (1.00%) high severe

Are there any user-facing changes?