Automatic Running of Tests On Pull Request

[saifrk]

**Fixes #1111 **

[saifrk]

A centralized logic has been implemented to streamline test execution across all workflows.
Continuous Execution:
Even if a test fails, the check doesn’t fail immediately.
It continues running other tests and, at the end, reports the total number of failed tests while marking the check as failed.
Naming Convention:
All tests beginning with tests_* are executed.
To add new tests, simply adhere to this naming convention.
Workflow Trigger

Workflow: Django Test Suite on PR
Trigger: Automatically triggered whenever a pull request (PR) is created or updated for the dev branch.
Workflow Steps

Checkout Merged Code:
The workflow checks out the merged code which combines:
The branch being merged .
The target branch (dev).
Purpose: Tests are executed on the post-merge state to ensure stability.
Set Up Environment:
Installs docker-compose.
Builds the Docker environment defined in local.yml.
Run Tests:
Executes the Django test suite which runs all test cases defined in the merged code.
Cleanup:
Cleans up Docker containers and volumes to free resources after tests finish.
Test Execution

1.Test Scope:

Executes all pytest tests defined in the repository.
Includes all files matching the naming convention tests_*.
2.Fail-Safe Mechanism:
Even if a test fails, all tests are executed.
At the end, it reports the number of failed tests and marks the workflow as failed if any test fails.
3.Feedback:
Results appear directly in the pull request as:
✅ All tests pass.
❌ Test failures are displayed with logs for debugging.
Key Change:

Previously, tests were executed on the branch being pushed, Integration issues with the target branch (e.g., dev) were not accounted for.
Now, Tests are executed on the merged code, this ensures the merged state is stable and prevents surprises post-merge.

[CarsonDavis]

Take a look at whether you can use https://github.com/NASA-IMPACT/COSMOS/blob/dev/pytest.ini as a way to handle what is being done in init.sh.

[saifrk]

The issue with using pytest.ini is that the Django’s URL namespaces sometimes don’t fully register when running all tests together, leading to NoReverseMatch errors. The init.sh script avoids this issue by running each test file separately, ensuring that the URL configuration is properly loaded every time.This approach is more consistent and reliable.

Note: Also resolved test failures in TestReindexingStatusTransitions by updating the mock to include the 'CollectionSelection' tag, ensuring proper execution of add_to_public_query()

[CarsonDavis]

Can you take a look and see what this chunk of the code is actually doing so that I can understand it better? It doesn't seem like it interacts with the other test cases.