Filtering dictionary data types using dplyr in Arrow package

Is there any way where we can filter the dictionary data types using dplyr. Below is the example for illustration 

test %>% dplyr::filter(cols %in% c("A","B","C") %>collect(). It gives me following error

Error in dataset___Scanner__ToTable(self) :   NotImplemented: Function is_in has no kernel matching input types (array[dictionary])

Here cols is dictionary type value

**Reporter**: [Soumya Ranjan Prusty](https://issues.apache.org/jira/secure/ViewProfile.jspa?name=prustys)

_{**Note**: *This issue was originally created as [PARQUET-1962](https://issues.apache.org/jira/browse/PARQUET-1962). Please see the [migration documentation](https://issues.apache.org/jira/browse/PARQUET-2502) for further details.*}

首先使用 dictionary 类型的 cols 值复现 dplyr 过滤示例，并检查报告没有匹配的 is_in 内核的 Arrow 处理逻辑。使用列出的值过滤 dictionary 数据类型时能够在没有 NotImplemented 错误的情况下完成，即表示完成。

由索引模型根据 Issue 内容生成。