feat: skip SQLAlchemy model fields with default values

[AlexPetul]

Description

Skip random value generation for SQLAlchemy fields that have a default value.

Although this works for most common cases, it has limitations for more advanced usage.
For example, context-sensitive default functions cannot be reliably evaluated:

def mydefault(context):
    return context.get_current_parameters()["counter"] + 12


t = Table(
    "mytable",
    metadata_obj,
    Column("counter", Integer),
    Column(
        "counter_plus_twelve",
        Integer,
        default=mydefault,
        onupdate=mydefault,
    ),
)

Such defaults require an execution context, and there is no obvious or safe way to construct a DefaultExecutionContext.

Similarly, SQL expressions used as defaults (for example text("select ...")) cannot be evaluated without a db round-trip, so they cannot be resolved at factory/build time.

This implementation only supports simple scalar defaults and callable defaults without context parameters.

Closes #837

[adhtruong]

Currently, the pattern used in other factories is to just detect that there is a default and omit supplying a value when this setting is on.

Does logic work if persistence is used? This may not be less complex though don't feel strongly on this given does feel clunkier to use

[AlexPetul]

Currently, the pattern used in other factories is to just detect that there is a default and omit supplying a value when this setting is on.

Indeed, dataclasses, msgspec, etc. handle defaults at the runtime level, so omitting a value applies the default. But with alchemy it's different.

Behavior with persistence remains unchanged:

• default is applied by SQLAlchemy before insert
• server_default is applied by the database if the column is omitted