Remove the zxJDBC dialect and its entry points

[aminghadersohi]

Problem

ibm_db_sa/zxjdbc.py imports sqlalchemy.connectors.zxJDBC, which SQLAlchemy removed in 1.4. This package declares four sqlalchemy.dialects entry points that point at it:

• db2.zxjdbc, db2.zxjdbc400
• ibm_db_sa.zxjdbc, ibm_db_sa.zxjdbc400

On any supported SQLAlchemy 1.4 or 2.0 install, loading them always raises ModuleNotFoundError: No module named 'sqlalchemy.connectors.zxJDBC'. The dialect also only ever worked on Jython, and Jython cannot install this package (python_requires >= 3.9), so it can't be used even with SQLAlchemy 1.3.

Applications that enumerate installed dialects through entry points, for example to show which databases are available, hit and log this error on every scan.

Change

• Remove ibm_db_sa/zxjdbc.py and its four entry points.
• Remove the two zxjdbc registrations from run_tests.py.
• README: known limitation 7 now mentions PyODBC only.

The ibm_db, pyodbc and AS/400 dialects, and all their entry points, are unchanged.

Verification

Python 3.11, SQLAlchemy 2.0.52. The script loads every sqlalchemy.dialects entry point this distribution declares:

Build	Loaded	Failed
master 7f3866e (0.4.4)	8	4 (all four zxjdbc: No module named 'sqlalchemy.connectors.zxJDBC')
this branch	8	0

The same four failures reproduce with released 0.4.0 on SQLAlchemy 1.4.54. The built wheel's entry_points.txt lists exactly the eight remaining dialects, and the wheel contains no zxjdbc module.