Add support for serialization of iceberg tables

[bolkedebruin]

This adds lightweight support for serialization of Apache Iceberg tables. This means that references
are captured and tables are reinstantiated with their catalog information.

Lightweight means that Catalogs are not reinstated from connections-ids or more generically. I think that should find its home in airflow.catalog but that is up for discussion.

cc @Fokko

This allows you to do the following:

import polars as pl
from pyiceberg.catalog import load_catalog
from pyiceberg.table import Table

@task
def a_task() -> Table:
  catalog = load_catalog("airflow")
  table = catalog.load_table("schema.table")
  return table

@task
def b_task(table: Table) -> pl.DataFrame:
  df = pl.scan_iceberg(table)
  return df

---

^ Add meaningful description above
Read the Pull Request Guidelines for more information.
In case of fundamental code changes, an Airflow Improvement Proposal (AIP) is needed.
In case of a new dependency, check compliance with the ASF 3rd Party License Policy.
In case of backwards incompatible changes please leave a note in a newsfragment file, named {pr_number}.significant.rst or {issue_number}.significant.rst, in newsfragments.

[hussein-awala]

Looks good!

I wonder if we should move some of the serializers to the providers (Kubernetes, this one, ...) and load them from there:

airflow/airflow/serialization/serde.py

Line 339 in 30ddfc5

for _, name, _ in iter_namespace(airflow.serialization.serializers):

But until we do that, LGTM.

[bolkedebruin]

I think so.

Looks good!

I wonder if we should move some of the serializers to the providers (Kubernetes, this one, ...) and load them from there:

airflow/airflow/serialization/serde.py

Line 339 in 30ddfc5

for _, name, _ in iter_namespace(airflow.serialization.serializers):

But until we do that, LGTM.