feat: Build for Python 3.14

[mattkjames7]

Description

This PR updates gqlalchemy for Python 3.14 readiness by combining CI/dependency compatibility updates with the previously merged Pydantic v2 API migration #384. It broadens test coverage across newer Python versions, adjusts optional dependency markers for unsupported stacks (notably TensorFlow/TF-GNN on newer interpreters), and keeps OGM behavior stable for normal Node/Relationship usage.

Most users should not need code changes. One edge case remains: nullable (Optional[...]) fields may accept omitted values as None where strict “required but nullable” validation was previously relied on.

Added a short smoke-test of gqlalchemy in macOS and restored tests for Windows in CI.

Pull request type

Please delete options that are not relevant.

• Bugfix
• Feature
• Code style update (formatting, renaming)
• Refactoring with functional or API changes
• Refactoring without functional or API changes
• Build or packaging related changes
• Documentation content changes
• Other (please describe):

Related issues

Delete section if this PR doesn't resolve any issues.

Closes (link to issue)

Checklist:

• My code follows the style guidelines of this project
• I have performed a self-review of my own code
• I have made corresponding changes to the documentation
• My changes generate no new warnings
• I have added tests that prove my fix is effective or that my feature works
• New and existing unit tests pass locally with my changes

######################################

Reviewer checklist (the reviewer checks this part)

• Core feature implementation
• Tests
• Code documentation
• Documentation on gqlalchemy/docs

######################################

[Copilot]

Pull request overview

This PR targets Python 3.14 readiness by updating CI matrices and dependency markers, and aligning GQLAlchemy’s OGM internals/tests with Pydantic v2 APIs while preserving historical OGM behavior (notably around omitted nullable fields).

Changes:

• Expand CI coverage up through Python 3.14 and adjust test selection/extras installation per interpreter.
• Update OGM internals and tests to use Pydantic v2 APIs (model_fields, model_validate, field_validator) and add compatibility helpers for field metadata.
• Refine dependency markers for optional stacks (TensorFlow/TF-GNN, PyArrow) and update docs to reflect modern Python support and Pydantic-v2 usage.

Reviewed changes

Copilot reviewed 17 out of 18 changed files in this pull request and generated 7 comments.

Show a summary per file

File	Description
tests/transformations/translators/test_tfgnn_transformations.py	Switch TF/TF-GNN imports to importorskip and add tfgnn markers.
tests/ogm/test_validators.py	Migrate validators to Pydantic v2 field_validator and adjust list validation logic.
tests/ogm/test_properties.py	Replace __fields__ assertions with model_fields.
tests/ogm/test_loading.py	Use Pydantic v2 ValidationError import.
tests/ogm/test_custom_fields.py	Use GQLAlchemy Field wrapper instead of Pydantic v1 Field.
tests/ogm/test_class_definition.py	Replace __fields__ assertions with model_fields.
tests/docs/test_ogm.py	Add module-level skip for Python 3.14+ due to behavior mismatch.
pyproject.toml	Add Python-version markers for pandas/TF stacks and update optional extras markers.
gqlalchemy/vendors/memgraph.py	Update on-disk property handling to use model-field helpers/attrs.
gqlalchemy/models.py	Implement Pydantic v2 migration helpers, Field wrapper metadata, and model_validate flow.
gqlalchemy/connection.py	Switch parse_obj usages to model_validate.
gqlalchemy/__init__.py	Update exported validator import to Pydantic v2 location.
docs/reference/gqlalchemy/models.md	Document parse_obj compatibility and recommend model_validate.
docs/installation.md	Update supported Python statement and clarify optional-extras availability.
docs/index.md	Update support note to include Python 3.14 and extras caveats.
docs/how-to-guides/ogm.md	Add note about omitted nullable fields being treated as None.
.github/workflows/build-and-test.yml	Expand Python matrices, add dispatch inputs, split dependency installs and test runs by Python version.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[antejavor]

Massive improvements all around the places (python support, win, macos build) , with a minor downside in compatibility.

I guess at some point we will ditch most of the dependencies out and move to Memgraph QM ->GNN/TGNN, being an example, since we now have this ->https://github.com/memgraph/memgraph/blob/master/mage/python/gnn.py

That means dependencies can live inside the mage container, keeping clients cleaner and leaner. 🦾