Add unit tests for LSS master (coverage 42% → 93%)

[bizfsc]

Summary

Add 28 unit tests for canopen/lss.py to improve test coverage from 42% to 93%.

Tests Added

• Switch state commands: global configuration/waiting mode, selective with match/no-match
• Node ID configuration: set node ID, error response handling
• Bit timing configuration: set bit timing, activate with delay
• Store configuration: success and error responses
• LSS address inquiry: vendor ID, product code, revision number, serial number
• Timeout handling: verify LssError raised on missing responses
• Fast scan: full protocol flow with binary search
• Obsolete aliases: send_switch_state_global backward compat
• LssError: exception instantiation and inheritance

Approach

Uses unittest.mock.MagicMock to mock the network layer. A helper method injects CAN responses during send_message() calls to simulate real LSS slave behavior, including only responding to commands that expect a reply per the protocol.

Coverage

canopen/lss.py    42% → 93%
Overall           77% → 80%

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

[acolomb]

Good stuff, thank you!

As with other modules' tests, the raw byte sequence should be compared to catch wrong constants definitions, instead of using those from the module to be tested. I reworked that accordingly and fixed some style nits. Few comments remain, please have a look.

[bizfsc]

Thanks for reviewing, you are 100% correct and I also like the suggested import-style more.
Disclaimer: I used Claude Opus 4.6 to generate those tests after analysis and reviewed them manually.

[acolomb]

I used Claude Opus

So I guessed. :-) But that's fine with me as long as the result is sane and reviewed by yourself.