feat: add worker kwargs

[leoparente]

This pull request introduces significant improvements to the policy runner's backend integration and job scheduling, focusing on supporting backend caching, forwarding additional arguments, and enhancing test coverage. It also includes minor workflow adjustments for linting jobs.

Backend extensibility and caching:

• The Backend.run method now accepts arbitrary keyword arguments (**kwargs), allowing the policy runner to pass additional parameters such as cache objects and scheduling controls. ([worker/worker/backend.pyL28-R36](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-3b735cc0006b8dac829eecfb9a8832393219518c12aaed7535e4410a5c36ab10L28-R36))
• The policy runner (PolicyRunner) now supports a backend cache mechanism: it attempts to create a cache using a backend's create_cache method if available, or falls back to cache config or an empty dict. The cache is injected into backend runs and reset on runner stop. ([[1]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2R37-R39), [[2]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2R58-R61), [[3]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2R83-R92), [[4]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2R254-R264), [[5]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2R200-R223))
• The runner also forwards schedule_now and any other relevant configuration to backend jobs, making the job execution context more flexible. ([[1]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2R83-R92), [[2]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2L94-R110))

Job scheduling and execution improvements:

• The runner maintains internal job args and kwargs, uses them for scheduled and immediate jobs, and provides a schedule_now method to trigger immediate backend runs with merged kwargs. ([[1]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2R37-R39), [[2]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2R83-R92), [[3]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2L94-R110), [[4]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2R200-R223))
• The run method ensures the cache is always passed to the backend, either from the runner or from explicit kwargs. ([[1]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2L104-R124), [[2]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-77ccd0c70b117c80a734aad5873b3fce2bfca6d5d1486c8bcf12f6e19cdd57e2R141-R145))

Testing enhancements:

• Extensive new and updated tests verify that caches and extra kwargs are forwarded correctly, that backend cache factories are used, and that immediate scheduling works as intended. Tests for error handling and metrics are also updated to reflect the new cache logic. ([[1]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-f6ad41627bb040e8e4b7e13d214a73c2bd73b5da5f9882ec7bfe9b1c83c63a51R114-R121), [[2]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-f6ad41627bb040e8e4b7e13d214a73c2bd73b5da5f9882ec7bfe9b1c83c63a51R144-R150), [[3]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-f6ad41627bb040e8e4b7e13d214a73c2bd73b5da5f9882ec7bfe9b1c83c63a51R174-R255), [[4]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-f6ad41627bb040e8e4b7e13d214a73c2bd73b5da5f9882ec7bfe9b1c83c63a51L180-R305), [[5]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-f6ad41627bb040e8e4b7e13d214a73c2bd73b5da5f9882ec7bfe9b1c83c63a51R315), [[6]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-f6ad41627bb040e8e4b7e13d214a73c2bd73b5da5f9882ec7bfe9b1c83c63a51L215-R336), [[7]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-f6ad41627bb040e8e4b7e13d214a73c2bd73b5da5f9882ec7bfe9b1c83c63a51R353), [[8]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-f6ad41627bb040e8e4b7e13d214a73c2bd73b5da5f9882ec7bfe9b1c83c63a51L241-R379), [[9]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-f6ad41627bb040e8e4b7e13d214a73c2bd73b5da5f9882ec7bfe9b1c83c63a51R408-R409), [[10]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-f6ad41627bb040e8e4b7e13d214a73c2bd73b5da5f9882ec7bfe9b1c83c63a51L302-R432), [[11]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-d55e46e202e1443376814c891586fe105d25509f22b905c49f66421d6818fc31R39-R51), [[12]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-759b8898f054470a76ad681c97fa2375588f74cd9cfc60bd7a633854ecf2af78R76-R80))

Workflow/linting:

• The GitHub Actions workflows for device discovery and worker linting are tightened by removing continue-on-error: true, making lint failures blocking. ([[1]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-5825ca970a22eff6f4d1f512a9fcf8b6dd6cc38e2a132379c39062ad2482ababL61), [[2]](https://github.com/netboxlabs/orb-discovery/pull/192/files#diff-cc57007a672d909cc9b067a40e7402b56b05e647ae6aaab926f4bf8b9115a129L61))

These changes make the policy runner more robust, extensible, and testable, especially for backends that need persistent state or custom job parameters.

[github-actions]

Coverage Report

File	Stmts	Miss	Cover	Missing
worker
main.py	54	4	93%	185–186, 195, 201
metrics.py	53	1	98%	104
server.py	81	4	95%	34–36, 167, 170
version.py	7	1	86%	14
worker/policy
manager.py	48	3	94%	36–37, 52
runner.py	128	5	96%	117, 211, 263–265
TOTAL	428	18	96%

Tests	Skipped	Failures	Errors	Time
63	0 💤	0 ❌	0 🔥	2.581s ⏱️