[VL] try_cast does not return NULL

### Backend

VL (Velox)

### Bug description

```
val data = Seq(("1234567"), ("000A1234"), ("ABCDE0"))
val df = spark.createDataFrame(data.map(Tuple1(_))).toDF("value")
df.write.format("parquet").mode("overwrite").save("file:///home/a/src/sparklocal/tbl)
spark.read.format("parquet").load("file:///home/a/src/sparklocal/wrong").createOrReplaceTempView("tbl")
val df = sql("select try_cast(value as bigint) from tbl")
```
Returns
```
E1126 15:27:22.105729 44701 CastExpr-inl.h:43] Cannot cast VARCHAR '000A1234' to BIGINT. Encountered a non-digit character
I1126 15:27:22.106130 44701 Task.cpp:1879] Terminating task Gluten_Stage_2_TID_2_VTID_1 with state Failed after running for 5 ms.
I1126 15:27:22.106395 44701 Task.cpp:1136] All drivers (1) finished for task Gluten_Stage_2_TID_2_VTID_1 after running for 5 ms.
24/11/26 15:27:22 ERROR TaskResources: Task 2 failed by error: org.apache.gluten.exception.GlutenException: org.apache.gluten.exception.GlutenException:
Exception: VeloxUserError
Error Source: USER
Error Code: INVALID_ARGUMENT
Reason: Cannot cast VARCHAR '000A1234' to BIGINT. Encountered a non-digit character
Retriable: False
Context: Top-level
Expression: cast((trim(
     　       

:VARCHAR, n0_0)) as BIGINT)
```
While Spark returns Array([1234567], [null], [null])

### Spark version

Spark-3.5.x

### Spark configurations

_No response_

### System information

_No response_

### Relevant logs

_No response_

Start by reproducing the Spark SQL query using Velox and inspect the cast failure reported from CastExpr-inl.h:43. Compare try_cast behavior for valid and invalid strings with Spark; done means invalid BIGINT conversions return NULL and the query completes instead of failing the task.

Written by the indexing model from the issue text.