[BugFix] Fix BE crash in primary key auto-increment partial update apply (backport #76119)

[mergify]

Why I'm doing:

For a primary-key table with an AUTO_INCREMENT column, a partial update that
omits that column applies through SegmentRewriter::rewrite_auto_increment, which
rebuilds the full segment from three sources: the keys / user-provided columns
re-read from the partial segment, the auto-increment column, and the missing
columns read from history.

rewrite_auto_increment swallowed the status of the partial-segment read: the
iterator was taken with StatusOr::value() (no ok() check) and get_next's status
was dropped (only a DCHECK guarded the row count, compiled out in release). A
transient read failure left read_chunk short, so the rebuilt segment recorded a
segment num_rows of 0 while its value columns still held N rows of real data.
The rowset meta and primary index kept the commit-time counts, so the
inconsistency stayed dormant until a later partial update of the same keys read
that segment via get_column_values, which silently skipped the 0-row segment,
returned an empty read column, and dereferenced a null source in
append_selective / SIMDGather (src=0x0) -- a crash that replays the pending
apply on every restart.

Shared-nothing only; the lake path (rewrite_auto_increment_lake)
already checks the read.

What I'm doing:

Root cause (write side), be/src/storage/rowset/segment_rewriter.cpp:

• rewrite_auto_increment no longer swallows read errors: create the iterator
  with ASSIGN_OR_RETURN (an iterator-creation failure now returns cleanly
  instead of aborting in value()), propagate get_next's status, and always
  validate the read row count against the segment, with the expected/actual
  counts and segment path in the error. This mirrors rewrite_auto_increment_lake.
  Also closes a latent crash where itr->close() ran on a possibly-null iterator.

Read side (stops the crash on already-corrupt data),
be/src/storage/tablet_updates.cpp:

• get_column_values returns Status::Corruption instead of silently skipping when
  a resolved rssid points to a 0-row segment (both the main read loop and the
  auto-increment default block). Corruption is non-retryable, so the tablet
  enters error state (recoverable via clone) rather than an endless crash-loop.
  Generalizes [BugFix] Guard against iterator UB in get_column_values when rssid not found #69617, which only guarded the rssid-underflow / out-of-range case.

Fixes #issue

What type of PR is this:

• BugFix
• Feature
• Enhancement
• Refactor
• UT
• Doc
• Tool

Does this PR entail a change in behavior?

• Yes, this PR will result in a change in behavior.
• No, this PR will not result in a change in behavior.

If yes, please specify the type of change:

• Interface/UI changes: syntax, type conversion, expression evaluation, display information
• Parameter changes: default values, similar parameters but with different default values
• Policy changes: use new policy to replace old one, functionality automatically enabled
• Feature removed
• Miscellaneous: upgrade & downgrade compatibility, etc.

Checklist:

• I have added test cases for my bug fix or my new feature
• This pr needs user documentation (for new or modified features or behaviors)
  • I have added documentation for my new feature or new function
  • This pr needs auto generate documentation
• This is a backport pr

Bugfix cherry-pick branch check:

• I have checked the version labels which the pr will be auto-backported to the target branch
  • 4.1
  • 4.0
  • 3.5

---

This is an automatic backport of pull request #76119 done by [Mergify](https://mergify.com).

[mergify]

Cherry-pick of ce03ba6 has failed:

On branch mergify/bp/branch-3.5/pr-76119
Your branch is up to date with 'origin/branch-3.5'.

You are currently cherry-picking commit ce03ba620f.
  (fix conflicts and run "git cherry-pick --continue")
  (use "git cherry-pick --skip" to skip this patch)
  (use "git cherry-pick --abort" to cancel the cherry-pick operation)

Changes to be committed:
	modified:   be/src/storage/rowset/segment_rewriter.cpp

Unmerged paths:
  (use "git add <file>..." to mark resolution)
	both modified:   be/src/storage/tablet_updates.cpp
	both modified:   be/test/storage/rowset/rowset_test.cpp
	both modified:   be/test/storage/tablet_updates_test.cpp

To fix up this pull request, you can check it out locally. See documentation: https://docs.github.com/en/pull-requests/collaborating-with-pull-requests/reviewing-changes-in-pull-requests/checking-out-pull-requests-locally

[mergify]

@mergify[bot]: Backport conflict, please reslove the conflict and resubmit the pr