perf: optimize left function by eliminating double chars() iteration (#19571)

For negative n values, the function was calling string.chars() twice:
1. Once to count total characters
2. Again to take the prefix

This optimization collects chars into a reusable buffer once per row for
the negative n case, eliminating the redundant iteration.

Benchmark results (negative n, which triggers the optimization):
- size=1024: 71.323 µs → 52.760 µs  (26.0% faster)
- size=4096: 289.62 µs → 212.23 µs  (26.7% faster)

Benchmark results (positive n, minimal overhead):
- size=1024: 24.465 µs → 24.691 µs  (0.9% slower)
- size=4096: 96.129 µs → 97.078 µs  (1.0% slower)

The dramatic improvement for negative n cases far outweighs the
negligible overhead for positive n cases.

## Which issue does this PR close?

<!--
We generally require a GitHub issue to be filed for all bug fixes and
enhancements and this helps us generate change logs for our releases.
You can link an issue to this PR using the GitHub syntax. For example
`Closes #123` indicates that this PR will close issue #123.
-->

- Closes #.

## Rationale for this change

<!--
Why are you proposing this change? If this is already explained clearly
in the issue then this section is not needed.
Explaining clearly why changes are proposed helps reviewers understand
your changes and offer better suggestions for fixes.
-->

## What changes are included in this PR?

<!--
There is no need to duplicate the description in the issue here but it
is sometimes worth providing a summary of the individual changes in this
PR.
-->

## Are these changes tested?

<!--
We typically require tests for all PRs in order to:
1. Prevent the code from being accidentally broken by subsequent changes
2. Serve as another way to document the expected behavior of the code

If tests are not included in your PR, please explain why (for example,
are they covered by existing tests)?
-->

## Are there any user-facing changes?

<!--
If there are user-facing changes then we may require documentation to be
updated before approving the PR.
-->

<!--
If there are any breaking changes to public APIs, please add the `api
change` label.
-->

---------

Co-authored-by: Claude Sonnet 4.5 <[email protected]>
Co-authored-by: Andrew Lamb <[email protected]>

Author: 3 people

datafusion/functions/src/unicode/left.rs

@@ -139,14 +139,23 @@ fn left_impl<'a, T: OffsetSizeTrait, V: ArrayAccessor<Item = &'a str>>(
     n_array: &Int64Array,
 ) -> Result<ArrayRef> {
     let iter = ArrayIter::new(string_array);
+    let mut chars_buf = Vec::new();
     let result = iter
         .zip(n_array.iter())
         .map(|(string, n)| match (string, n) {
             (Some(string), Some(n)) => match n.cmp(&0) {
                 Ordering::Less => {
-                    let len = string.chars().count() as i64;
-                    Some(if n.abs() < len {
-                        string.chars().take((len + n) as usize).collect::<String>()
+                    // Collect chars once and reuse for both count and take
+                    chars_buf.clear();
+                    chars_buf.extend(string.chars());
+                    let len = chars_buf.len() as i64;
+
+                    // For negative n, take (len + n) chars if n > -len (avoiding abs() which panics on i64::MIN)
+                    Some(if n > -len {
+                        chars_buf
+                            .iter()
+                            .take((len + n) as usize)
+                            .collect::<String>()
                     } else {
                         "".to_string()
                     })