Avro: read named type references and write unique fixed type names

[ozenyunus35]

Closes #4116

Rationale for this change

As described in the issue, an Avro named type (decimal_P_S, fixed_L, uuid_fixed) may only be defined once per schema; later uses must refer to it by name. PyIceberg couldn't read the reference form (TypeError: Type not recognized: decimal_10_2), so it failed on manifests written by Java and iceberg-rust, and it wrote the same definition every time a type was reused, which strict readers reject.

• Read: AvroSchemaConversion remembers each named type (record, enum, fixed) it converts, and resolves a later string type with that name to the same Iceberg type. A repeated full definition is still accepted, so existing manifests keep working. The map is reset at the start of avro_to_iceberg.
• Write: following the review, ConvertSchemaToAvro adds the field-id of the field, list element or map key/value that uses a fixed, decimal or uuid type to its name (decimal_10_2_1, fixed_3_5, uuid_fixed_7), the same way records are named r<field-id>. Field-ids are unique, so every definition has its own name and no bookkeeping of defined names is needed.

Are these changes tested?

Yes, in the existing test files:

• tests/utils/test_schema_conversion.py: reading the reference form (the repro from the issue, plus fixed_3 and uuid_fixed), reading the repeated form, an unknown name still raising TypeError, and writing a schema with repeated decimal, fixed (one inside a struct) and uuid types, including a list element and a map key and value: each gets the field-id in its name and the result reads back to the same schema.
• tests/avro/test_file.py: test_repeated_named_types_round_trip writes a file with two decimal(10,2) and two uuid fields through AvroOutputFile and reads it back with PyIceberg and with fastavro. Without the change, fastavro refuses the file with a SchemaParseException for the redefined name.

The new tests fail without the change. tests/utils/test_schema_conversion.py, tests/avro, tests/test_avro_sanitization.py and tests/utils/test_manifest.py pass. Failures in my environment from optional extras I didn't install are the same before and after the change. ruff 0.14.3 and mypy 1.18.2 (the pre-commit versions) are clean.

Are there any user-facing changes?

Manifests and manifest lists that refer to a named type by name can now be read, and the fixed, decimal and uuid types PyIceberg writes now have the field-id in their name.

I used an AI assistant to help draft parts of the code and tests; I reviewed the changes and ran the tests myself.

[Fokko]

I've suggested another approach in one of the comments, let me know what you think 👍

[Fokko]

Let's use a frozenset here

[ozenyunus35]

Done in 4f95b2e.

[Fokko]

I think we're doing quite a bit of additional bookkeeping. I'm wondering if it wouldn't be easier to include the field-id to the name (which is unique by definition).

[ozenyunus35]

Good idea, thanks. 4f95b2e drops the bookkeeping: a fixed, decimal or uuid type now gets the field-id of the field, list element or map key/value that uses it in its name (e.g. decimal_10_2_1), like records are named r<field-id>. Reading the reference form is unchanged.