✨ Add PEP 593 support for Requirement annotations

[bolu61]

Fixes #229 (see #229 (comment)).

I'm not really sure how Pydantic does it, but here's a potential workaround to the problem.

[YuriiMotov]

@bolu61, thanks for working on this!

So, this PR fixes the following code example:

class Foo(SQLModel, table=True):
    id: Annotated[int | None, Field(primary_key=True)] = None
    uuid: Annotated[UUID, Field(unique=True)]

class Bar(SQLModel, table=True):
    id: Annotated[int | None, Field(primary_key=True)] = None
    uuid: Annotated[UUID, Field(unique=True)]
    foo_id: Annotated[int, Field(foreign_key="foo.id")]
    foo: Annotated[Foo, Relationship()]

For now last line (with Relationship()) doesn't work (ValueError: <class '__main__.Foo'> has no matching SQLAlchemy type) and this PR fixes that.

I think this is not the same that was initially proposed in #229. The idea there was to support passing multiple annotations like:

class Hero(SQLModel, table=True):
    id: Annotated[Optional[int], Field(examples=....), Column(primary_key=True)] = None

Where Field can be pydantic.Field.
So, I suggest we unmark this PR as the one that resolves that issue.

As for changes introduced by this PR, I think it goes in line with modern recommendations to use Annotated approach instead parameterizing fields via default value.
I think we should test this approach more and add some automatic tests. Then it will have all chances to be accepted.