perf: Use get_unchecked in strpos for micro-optimization

Use `get_unchecked` to avoid redundant bounds checking when comparing
substring slices, as we already validate the bounds in the if condition.

Benchmark results show modest improvements, especially for shorter strings:
- strpos_StringArray_ascii_str_len_8: 3.7% faster
- strpos_StringViewArray_ascii_str_len_8: 2.0% faster
- strpos_StringViewArray_utf8_str_len_8: 1.9% faster

Other benchmarks show no significant change (within noise threshold).

Author: viirya

datafusion/functions/src/unicode/strpos.rs

@@ -232,11 +232,16 @@ where
                     let mut char_pos = 0;
                     for (byte_idx, _) in string.char_indices() {
                         char_pos += 1;
-                        if byte_idx + substring_bytes.len() <= string_bytes.len()
-                            && &string_bytes[byte_idx..byte_idx + substring_bytes.len()]
-                                == substring_bytes
-                        {
-                            return T::Native::from_usize(char_pos);
+                        if byte_idx + substring_bytes.len() <= string_bytes.len() {
+                            // SAFETY: We just checked that byte_idx + substring_bytes.len() <= string_bytes.len()
+                            let slice = unsafe {
+                                string_bytes.get_unchecked(
+                                    byte_idx..byte_idx + substring_bytes.len(),
+                                )
+                            };
+                            if slice == substring_bytes {
+                                return T::Native::from_usize(char_pos);
+                            }
                         }
                     }