feat: Add SQLAlchemy models for Chancy tables

[njg7194]

Summary

This PR adds SQLAlchemy models as a new contrib module, mirroring the existing Django models. This addresses #75.

Changes

• chancy/contrib/sqlalchemy/init.py - Module documentation
• chancy/contrib/sqlalchemy/models.py - SQLAlchemy models for Job, Worker, and Queue tables
• tests/contrib/sqlalchemy/test_models.py - Integration tests

Features

• Full SQLAlchemy 2.0 style with mapped_column and Mapped types
• PostgreSQL-specific types (JSONB, ARRAY, UUID)
• Table prefix customization via CHANCY_PREFIX environment variable
• Comprehensive docstrings with usage examples
• Unmanaged models (Chancy handles migrations)

Usage Example

from sqlalchemy import create_engine, select
from sqlalchemy.orm import Session
from chancy.contrib.sqlalchemy.models import Job, Queue, Worker

engine = create_engine("postgresql://...")

with Session(engine) as session:
    # Get all pending jobs
    pending_jobs = session.execute(
        select(Job).where(Job.state == "pending")
    ).scalars().all()
    
    # Get all active queues
    active_queues = session.execute(
        select(Queue).where(Queue.state == "active")
    ).scalars().all()

Closes #75

[PaulM5406]

Thanks a lot @njg7194 for the PR. I will let @TkTech review the implementation details more thoroughly. Reading it, I have a few small comments.

[PaulM5406]

nit: this documentation uses the former SQLA syntaxe, i.e. session.query.
test_models.py uses modern syntaxe session.execute though.

[PaulM5406]

nit: For the whole file, nullable can be inferred from the type.

Suggested change

	queue: Mapped[str] = mapped_column(Text, nullable=False)
	queue: Mapped[str] = mapped_column(Text)

If Optional is used, nullable=True is inferred.

[PaulM5406]

It is really useful ?