Schema::try_merge should be able to merge List of any data type with List of Null data type

[mnpw]

Describe the bug
Schema::try_merge should be able to merge List(any data type) and List(Null data type)

To Reproduce

use serde_json::json;

let o_1 = vec![
    json!({"a": [1,2,3]}),
    json!({"a": [4,5,6]}),
];

let o_2 = vec![
    json!({"a": []}),
    json!({"a": []}),
];

let o_3 = vec![
    json!({"a": []}),
    json!({"a": [7,8,9]}),
];

let schema_1 = arrow_json::reader::infer_json_schema_from_iterator(o_1.iter().map(Ok)).unwrap();
println!("{:?}", schema_1);
// Schema { fields: [Field { name: "a", data_type: List(Field { name: "item", data_type: Int64, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }), nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }], metadata: {} }

let schema_2 = arrow_json::reader::infer_json_schema_from_iterator(o_2.iter().map(Ok)).unwrap();
println!("{:?}", schema_2);
// Schema { fields: [Field { name: "a", data_type: List(Field { name: "item", data_type: Null, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }), nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }], metadata: {} }

let schema_3 = arrow_json::reader::infer_json_schema_from_iterator(o_3.iter().map(Ok)).unwrap();
println!("{:?}", schema_3);
// Schema { fields: [Field { name: "a", data_type: List(Field { name: "item", data_type: Int64, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }), nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }], metadata: {} }

let merged = Schema::try_merge([schema_1, schema_2, schema_3]);
println!("{:?}", merged)
// Err(SchemaError("Fail to merge schema field 'a' because the from data_type = List(Field { name: \"item\", data_type: Null, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }) does not equal List(Field { name: \"item\", data_type: Int64, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} })"))

Expected behavior
Merge of List( data type) and List(Null) should be possible

let merged = Schema::try_merge([schema_1, schema_2, schema_3]);
println!("{:?}", merged)
// Ok(Schema { fields: [Field { name: "a", data_type: List(Field { name: "item", data_type: Int64, nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }), nullable: true, dict_id: 0, dict_is_ordered: false, metadata: {} }], metadata: {} })

Additional context
N/A

[alamb]

label_issue.py automatically added labels {'arrow'} from #5852