Core : snatshot read with schema change

[zhangjun0x01]

When we modify the schema of the iceberg table and do a time travel query by specifying the snapshot-id, the query result shows the current latest schema, but we want the schema of query result to be corresponding with the snapshot.

I add a schema-id to the snapshot in metadata , when we query by snapshot-id , we get the schema by the schema-id in the snapshot

@rdblue @aokolnychyi @openinx could you help me review it ? thanks.

[zhangjun0x01]

I found that this pr seems to be duplicate. I found other similar pr (#2275 ,#2096). I will close this pr, but I think we should add test cases for different computing engines (spark, flink, hive) so that they can works correctly.