[BUG] ORC file count statistic for nested type is wrong

**Describe the bug**
The GPU ORC file statistics show that the count for nested type is wrong while the CPU ORC file is correct.

GPU file shows different counts for nested type:
GPU:
```
File Statistics:
Column 0: count: 8 hasNull: true
Column 1: count: 1 hasNull: true
```
CPU:
```
File Statistics:
Column 0: count: 8 hasNull: false
Column 1: count: 8 hasNull: false
```
The data in both files are:
```
+------------+
| struct_s|
+------------+
|{null, null}|
| {1, 1}|
|{null, null}|
| {3, 3}|
|{null, null}|
| {5, 5}|
|{null, null}|
| {7, 7}|
+------------+
```

**Steps/Code to reproduce bug**
##### Generate GPU file
```cpp
TEST_F(OrcWriterTest, NestedColumnSelection)
{
auto const num_rows = 8;
std::vector child_col1_data(num_rows);
std::vector child_col2_data(num_rows);
for (int i = 0; i < num_rows; ++i) {
child_col1_data[i] = i;
child_col2_data[i] = i;
}

auto validity = cudf::detail::make_counting_transform_iterator(0, [](auto i) { return i % 2; });
int32_col child_col1{child_col1_data.begin(), child_col1_data.end(), validity};
int32_col child_col2{child_col2_data.begin(), child_col2_data.end(), validity};
struct_col s_col{child_col1, child_col2};
cudf::table_view expected({s_col});

cudf::io::table_input_metadata expected_metadata(expected);
expected_metadata.column_metadata[0].set_name("struct_s");
expected_metadata.column_metadata[0].child(0).set_name("field_a");
expected_metadata.column_metadata[0].child(1).set_name("field_b");

auto filepath = "/tmp/test-count-for-nested-type-gpu.orc";
cudf::io::orc_writer_options out_opts =
cudf::io::orc_writer_options::builder(cudf::io::sink_info{filepath}, expected)
.metadata(std::move(expected_metadata));
cudf::io::write_orc(out_opts);
}
```

Read the GPU file
SPARK_HOME/bin/pyspark

spark.read.orc("/tmp/test-count-for-nested-type-gpu.orc").show()
+------------+
| struct_s|
+------------+
|{null, null}|
| {1, 1}|
|{null, null}|
| {3, 3}|
|{null, null}|
| {5, 5}|
|{null, null}|
| {7, 7}|
+------------+

##### Generate CPU file
SPARK_HOME/bin/pyspark

```python
from pyspark.sql.types import *
schema = StructType([StructField("struct_s",
StructType([
StructField("field_a", IntegerType()),
StructField("field_b", IntegerType()),
]))])

def get_value(i):
if i % 2 == 0:
return None
else:
return i

data = [
({ 'field_a': get_value(i), 'field_b': get_value(i) }, ) for i in range(0, 8)
]
df = spark.createDataFrame(
SparkContext.getOrCreate().parallelize(data, numSlices=1),
schema)

path = '/tmp/test-count-for-nested-type-cpu.orc'
df.coalesce(1).write.mode("overwrite").orc(path)
spark.read.orc(path).show()
```

```
+------------+
| struct_s|
+------------+
|{null, null}|
| {1, 1}|
|{null, null}|
| {3, 3}|
|{null, null}|
| {5, 5}|
|{null, null}|
| {7, 7}|
+------------+
```

##### print count statistic for GPU file
```
$ orc-tool meta test-count-for-nested-type-gpu.orc
Processing data file test-count-for-nested-type-gpu.orc [length: 360]
Structure for test-count-for-nested-type-gpu.orc
File Version: 0.12 with ORIGINAL by ORC Java
Rows: 8
Compression: SNAPPY
Compression size: 262144
Calendar: Julian/Gregorian
Type: struct>

Stripe Statistics:
Stripe 1:
Column 0: count: 8 hasNull: true
Column 1: count: 1 hasNull: true
Column 2: count: 4 hasNull: true min: 1 max: 7 sum: 16
Column 3: count: 4 hasNull: true min: 1 max: 7 sum: 16

File Statistics:
Column 0: count: 8 hasNull: true
Column 1: count: 1 hasNull: true
Column 2: count: 4 hasNull: true min: 1 max: 7 sum: 16
Column 3: count: 4 hasNull: true min: 1 max: 7 sum: 16

Stripes:
Stripe: offset: 3 data: 24 rows: 8 tail: 92 index: 70
Stream: column 0 section ROW_INDEX start: 3 length 7
Stream: column 1 section ROW_INDEX start: 10 length 11
Stream: column 2 section ROW_INDEX start: 21 length 26
Stream: column 3 section ROW_INDEX start: 47 length 26
Stream: column 2 section PRESENT start: 73 length 5
Stream: column 2 section DATA start: 78 length 7
Stream: column 3 section PRESENT start: 85 length 5
Stream: column 3 section DATA start: 90 length 7
Encoding column 0: DIRECT
Encoding column 1: DIRECT
Encoding column 2: DIRECT_V2
Encoding column 3: DIRECT_V2

File length: 360 bytes
Padding length: 0 bytes
Padding ratio: 0%
```

##### print count statistic for CPU file
```
$ orc-tool meta /tmp/test-count-for-nested-type-cpu.orc
Processing data file file:/tmp/test-count-for-nested-type-cpu.orc/part-00000-6b490836-0c65-4355-9d0e-fbaff96aec33-c000.snappy.orc [length: 388]
Structure for file:/tmp/test-count-for-nested-type-cpu.orc/part-00000-6b490836-0c65-4355-9d0e-fbaff96aec33-c000.snappy.orc
File Version: 0.12 with ORC_14 by ORC Java 1.7.4
Rows: 8
Compression: SNAPPY
Compression size: 262144
Calendar: Julian/Gregorian
Type: struct>

Stripe Statistics:
Stripe 1:
Column 0: count: 8 hasNull: false
Column 1: count: 8 hasNull: false
Column 2: count: 4 hasNull: true bytesOnDisk: 12 min: 1 max: 7 sum: 16
Column 3: count: 4 hasNull: true bytesOnDisk: 12 min: 1 max: 7 sum: 16

File Statistics:
Column 0: count: 8 hasNull: false
Column 1: count: 8 hasNull: false
Column 2: count: 4 hasNull: true bytesOnDisk: 12 min: 1 max: 7 sum: 16
Column 3: count: 4 hasNull: true bytesOnDisk: 12 min: 1 max: 7 sum: 16

Stripes:
Stripe: offset: 3 data: 24 rows: 8 tail: 71 index: 76
Stream: column 0 section ROW_INDEX start: 3 length 11
Stream: column 1 section ROW_INDEX start: 14 length 11
Stream: column 2 section ROW_INDEX start: 25 length 27
Stream: column 3 section ROW_INDEX start: 52 length 27
Stream: column 2 section PRESENT start: 79 length 5
Stream: column 2 section DATA start: 84 length 7
Stream: column 3 section PRESENT start: 91 length 5
Stream: column 3 section DATA start: 96 length 7
Encoding column 0: DIRECT
Encoding column 1: DIRECT
Encoding column 2: DIRECT_V2
Encoding column 3: DIRECT_V2

File length: 388 bytes
Padding length: 0 bytes
Padding ratio: 0%

User Metadata:
org.apache.spark.version=3.3.0
```

**Expected behavior**
The all statistics should be correct, including the `hasNull`, refer to this [issue](https://github.com/rapidsai/cudf/issues/13817)

**Environment details**
Environment details
cuDF 23.08 branch
Spark 3.3.0
orc-core-1.7.4.jar

**Additional context**