SR API coverage still not quite complete when it comes to soft/hard delete

Hi,

there are still two more issues wrt SR API coverage when it comes to soft/hard delete.

1) get_subjects()
Does not support the boolean "deleted" parameter. In the SR API, you can get a combined list of non-deleted subjects and soft-deleted ones by adding "?deleted=True". This is not yet supported in the Python wrapper.

2) delete_versions()
It does support the parameter boolean "permanent", but there seems to be a bug in the internal caching. When I first delete a schema version (soft) and then try to trigger a hard-delete, I end up with:

`Traceback (most recent call last):
File "/opt/homebrew/Cellar/python@3.13/3.13.1/Frameworks/Python.framework/Versions/3.13/lib/python3.13/unittest/case.py", line 58, in testPartExecutor
yield
File "/opt/homebrew/Cellar/python@3.13/3.13.1/Frameworks/Python.framework/Versions/3.13/lib/python3.13/unittest/case.py", line 651, in run
self._callTestMethod(testMethod)
~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^
File "/opt/homebrew/Cellar/python@3.13/3.13.1/Frameworks/Python.framework/Versions/3.13/lib/python3.13/unittest/case.py", line 606, in _callTestMethod
if method() is not None:
~~~~~~^^
File "/Users/m0724822/kafka/kafi/test/test_single_storage_base.py", line 1278, in test_sr_two_versions
version_int = s.delete_version(key_subject_name_str, 1, permanent=True)
File "/Users/m0724822/kafka/kafi/kafi/schemaregistry.py", line 151, in delete_version
schema_id_int = self.schemaRegistryClient.delete_version(subject_name_str, version_int, permanent=permanent_bool)
File "/Users/m0724822/kafka/kafi/venv/lib/python3.13/site-packages/confluent_kafka/schema_registry/schema_registry_client.py", line 976, in delete_version
self._cache.remove_by_subject_version(subject_name, version)
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^
File "/Users/m0724822/kafka/kafi/venv/lib/python3.13/site-packages/confluent_kafka/schema_registry/schema_registry_client.py", line 509, in remove_by_subject_version
del self.rs_schema_index[subject][schema_id]
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^
KeyError: 5`