Feature Request: Add Support for EFCore Schemas by Treating Them as Databases

[Felixzed]

Feature Request

EFCore Entity Schemas can be treated like Clickhouse Databases to make EFCore output valid migrations.
In this feature request I propose a change to how Schemas are handled. I will be submitting a pull request along with my feature request.

Current behavior

Defining an EFCore Entity Configuration such that:

e.ToTable("my_table", "my_schema", table =>
{
	table.HasMergeTreeEngine()
		.WithOrderBy("Timestamp")
		.WithPartitionBy("toYear(Timestamp)");
});

		e.Property(x => x.Timestamp)
			.HasColumnType("Date")
			.HasColumnName("Timestamp");
		e.Property(x => x.MyString)
			.HasColumnType("String")
			.HasColumnName("MyString");
		e.HasNoKey();

and using that entity configuration in an EFCore migration produces the following migration code:

migrationBuilder.EnsureSchema(
    name: "my_schema");

migrationBuilder.CreateTable(
    name: "my_table",
    schema: "my_schema",
    columns: table => new
    {
        Timestamp = table.Column<DateOnly>(type: "Date32", nullable: false)
    },
    constraints: table =>
    {
    })
    .Annotation("ClickHouse:Engine", "MergeTree")
    .Annotation("ClickHouse:OrderBy", new[] { "Timestamp" })
    .Annotation("ClickHouse:PartitionBy", new[] { "toYear(Timestamp)" });

Which fails on the ensure schema operation, saying schemas are not supported. This makes it difficult define database structure in EFCore.

Expected behavior

// C# EFCore Migration
migrationBuilder.EnsureSchema(
    name: "pre_aggregates");

migrationBuilder.CreateTable(
    name: "my_table",
    schema: "my_schema",
    columns: table => new
    {
        Timestamp = table.Column<DateOnly>(type: "Date32", nullable: false)
         MyString = table.Column<string>(type: "String", nullable: false),

    },
    constraints: table =>
    {
    })
    .Annotation("ClickHouse:Engine", "MergeTree")
    .Annotation("ClickHouse:OrderBy", new[] { "Timestamp" })
    .Annotation("ClickHouse:PartitionBy", new[] { "toYear(Timestamp)" });

// Should translate to:

CREATE DATABASE my_schema;

CREATE TABLE `my_schema`.`my_table` (                                                                                                                                                                                                                                                                                                                                                                                    
    `Timestamp` Date32,
    `MyString` String                                                                                                                                                                                                                 
)                                                                                                                                                                                                                                           
ENGINE = MergeTree()                                                                                                                                                                                                             
ORDER BY (`Timestamp`)                                                                                                                                                                                                                      
PARTITION BY toYear(Timestamp);