[fix](function) preserve sign for negative sub-hour TIMESTAMPTZ offsets

[Mryange]

What problem does this PR solve?

Problem Summary:

When BE renders TIMESTAMPTZ values, it derives the timezone sign from the truncated hour component.
For negative sub-hour offsets such as -00:30, integer division makes the hour component 0, so the
renderer emits +00:30 instead of -00:30.

This affects both constant-folded expressions and runtime values rendered from table data.

Root cause:

• Old logic:

int offset_hours = time_offset / 3600;
int offset_mins = (std::abs(time_offset) % 3600) / 60;
char sign = offset_hours >= 0 ? '+' : '-';

• For time_offset = -1800, offset_hours becomes 0, so the sign is rendered as +.

Fix:

• Determine the sign from the original second offset.
• Split hour and minute fields from the absolute offset value.
• Add a focused UT for -00:30 rendering.

Release note

None

Check List (For Author)

• Test

  • Regression test
  • Unit Test
  • Manual test (add detailed scripts or steps below)
  • No need to test or manual test. Explain why:
    • This is a refactor/code format and no logic has been changed.
    • Previous test can cover this change.
    • No code files have been changed.
    • Other reason

• Behavior changed:

  • No.
  • Yes.

• Does this need documentation?

  • No.
  • Yes.

Check List (For Reviewer who merge this PR)

• Confirm the release note
• Confirm test cases
• Confirm document
• Add branch pick label

[Mryange]

run buildall

[hello-stephen]

Thank you for your contribution to Apache Doris.
Don't know what should be done next? See How to process your PR.

Please clearly describe your PR:

1. What problem was fixed (it's best to include specific error reporting information). How it was fixed.
2. Which behaviors were modified. What was the previous behavior, what is it now, why was it modified, and what possible impacts might there be.
3. What features were added. Why was this function added?
4. Which code was refactored and why was this part of the code refactored?
5. Which functions were optimized and what is the difference before and after the optimization?

[hello-stephen]

BE UT Coverage Report

Increment line coverage 100.00% (8/8) 🎉

Increment coverage report
Complete coverage report

Category	Coverage
Function Coverage	53.33% (20401/38255)
Line Coverage	36.87% (192169/521249)
Region Coverage	33.19% (149521/450450)
Branch Coverage	34.30% (65399/190648)

[hello-stephen]

BE Regression && UT Coverage Report

Increment line coverage 100.00% (8/8) 🎉

Increment coverage report
Complete coverage report

Category	Coverage
Function Coverage	71.53% (26818/37494)
Line Coverage	53.87% (280023/519860)
Region Coverage	47.26% (215474/455963)
Branch Coverage	50.55% (97504/192877)