feat: add GradeEventContextEnricher pipeline step for grade analytics

[pwnage101]

Introduces enterprise/filters/grades.py with GradeEventContextEnricher, a PipelineStep for the org.openedx.learning.grade.context.requested.v1 filter that enriches grade analytics event context with the learner's enterprise UUID. Adds openedx-filters as a dependency and unit tests covering both enrichment and no-op branches.

ENT-11563

Co-Authored-By: Claude Sonnet 4.6 noreply@anthropic.com

---

Blocked by:

• feat: add GradeEventContextRequested filter for grade analytics enrichment openedx-filters#333

Blocks:

• feat: replace enterprise_support import with GradeEventContextRequested filter openedx-platform#38097

[pwnage101]

Aside from these changes, please also remember to bump the version of edx-enterprise.

Apologies if this wasn't clear, but these draft claude-generated PRs were meant to offer guidance and direction, not to be ready to go as-is.

[pwnage101]

Add python type hints

Suggested change

	def run_filter(self, context, user_id, course_id): # pylint: disable=arguments-differ
	def run_filter(self, context: dict, user_id: int, course_id: str) -> dict[str, Any]: # pylint: disable=arguments-differ

[pwnage101]

Per the filter definition, it's impossible for course_id to be a CourseKey: https://github.com/openedx/openedx-filters/blob/d4b09fdb850436bbcea638066c124a47b5025084/openedx_filters/learning/filters.py#L1584

Fixed docstring:

Suggested change

	course_id (str or CourseKey): the course key for the grade event.
	course_id (str): the course key for the grade event.

[pwnage101]

Be more specific about the shape of the output dict.

Suggested change

	dict: updated pipeline data with the enriched ``context`` dict.
	dict: updated pipeline data with the enriched ``context`` dict.
	{
	"context": <enriched context>,
	"user_id": <unchanged>,
	"course_id": <unchanged>,
	}

[pwnage101]

Missing passthrough values and should probably just utilize updated_context

Suggested change

	return {"context": context}
	return {"context": context, "user_id": user_id, "course_id": course_id}

[pwnage101]

1. This lacks passthrough values, 2. probably shouldn't just be an early-return since that's less maintainable, and 3. leverages complicated destructuring instead of something more pythonic and something which allows us to retain one exit point of the function. Solve all 3 problems at once:

Suggested change

	return {"context": {**context, "enterprise_uuid": str(uuids[0])}}
	context = deepcopy(context) # create a copy to avoid altering the passed-in value.
	# Warning: Selecting the first element is not likely deterministic!
	context["enterprise_uuid"] = str(uuids[0])

[pwnage101]

❌ 3.3.0 now supports python 3.11. We should take this opportunity to deploy with the latest version.

Suggested change

	openedx-filters==2.1.0
	openedx-filters==3.3.0

Ditto for the rest of the requirements changes in this PR.

[pwnage101]

1. user_id can stay an int. I looked at the function, it's okay.
2. course_id can be assumed to be a str, that's promised by the type hint.

Suggested change

	uuids = EnterpriseCourseEnrollment.get_enterprise_uuids_with_user_and_course(
	str(user_id),
	str(course_id),
	)
	uuids = EnterpriseCourseEnrollment.get_enterprise_uuids_with_user_and_course(user_id, course_id)

[pwnage101]

1. Overly strict line length limits.
2. Unnecessary patching of module at import (enterprise.filters.grades) instead of source (enterprise.models).

Suggested change

	@patch(
	"enterprise.filters.grades.EnterpriseCourseEnrollment"
	".get_enterprise_uuids_with_user_and_course"
	)
	@patch("enterprise.models.EnterpriseCourseEnrollment.get_enterprise_uuids_with_user_and_course")

Ditto for other tests.

[pwnage101]

Use argument names to improve readability:

Suggested change

	return GradeEventContextEnricher(
	"org.openedx.learning.grade.context.requested.v1",
	[],
	)
	return GradeEventContextEnricher(
	filter_type="org.openedx.learning.grade.context.requested.v1",
	running_pipeline=[],
	)