[pipeline-connector][paimon] add paimon pipeline data sink connector.

[lvyanquan]

This close https://github.com/ververica/flink-cdc-connectors/issues/2856. Some codes are inspired by FlinkCdcMultiTableSink in Paimon repo, and add a sinkV2 implement.

[lvyanquan]

@PatrickRen PTAL.

[yanghuaiGit]

paimon在schemaschemachangevent事件产生时会从catalog里加载最新的schema，这个时候schema有可能没被修改，导致写入的数据还是修改ddl之前的schema字段数据，新的字段数据读不出来或者删除字段后出现新的问题 是否可以在releasestream之后发送schemaChangeEvebt事件，这样下游获取schema就一定是最新的

[lvyanquan]

Thanks @yanghuaiGit for pointing out this, address it.

[yanghuaiGit]

com.ververica.cdc.connectors.paimon.sink.PaimonMetadataApplier 静态字段catalog，在反序列化之后，获取的对象里catalog为null，导致com.ververica.cdc.connectors.paimon.sink.PaimonMetadataApplier#applySchemaChange方法执行时为空指针。

catalog应改为 private transient Catalog catalog;，在applySchemaChange时判断是否为空来构建一个catalog

[lvyanquan]

address it.

[melin]

Support reading data from multiple table messages written to the same topic？

https://github.com/apache/flink-cdc/pull/2938#issuecomment-1970940065

[lvyanquan]

Can multiple table cdc message be written to the same topic？

You can do this by using route in pipeline.

[yanghuaiGit]

paimon latest version is 0.7，we should update paimon version from 0.6 to 0.7

[PatrickRen]

paimon latest version is 0.7，we should update paimon version from 0.6 to 0.7

@lvyanquan Could you take a look at this one? I prefer to catch up with the latest version as well. Also could you rebase the latest master? Thanks

[lvyanquan]

paimon latest version is 0.7，we should update paimon version from 0.6 to 0.7

Done and rebased to master.

[melin]

Does kafka header set constant values? For example, if data from multiple data centers is written to the same kafka topic, add a region key to the kafka header.

[lvyanquan]

Thanks @yuxiqian for those comments, I've addressed it and resubmitted.