pimd: Fix Register-Stop state machine logic to align with RFC7761

[hhubb22]

The previous implementation of the PIM DR Register state machine did not immediately transition to Join-Pending after the suppression timer expired. This caused a 5-second "deaf period" where new Register-Stop messages were ignored.

This patch corrects the logic by:

• Updating the state to PIM_REG_JOIN_PENDING immediately upon suppression timer expiry, as per RFC7761 Figure 1.
• Ensuring Register-Stop messages are handled correctly in the Join-Pending state.
• Fix a redundant Null-Register transmission.

This change makes the PIM-SM implementation more robust and compliant.

[hhubb22]

Currently, FRR's PIM behavior when the DR receives a Register-Stop message does not conform to the state machine defined in RFC7761.
RFC7761 specifies that if the current state is Prune, when the register stop timer expires, the system should both send a null register and change the state to Register-Pending. However, the current FRR implementation misses this step, causing FRR to ignore and not process Register-Stop messages from the RP when it should actually respond to them.

[hhubb22]

ci:rerun

[Jafaral]

Fix style errors reported by frrbot please.

[hhubb22]

Fix style errors reported by frrbot please.

Okay, I have completed the style check for frrbot

[Jafaral]

Okay, I have completed the style check for frrbot

Thanks. No need to split the style fixes into a new commit. It is better to just squash that into the previous commit and pretend that you never had style issues ;).

[hhubb22]

Okay, I have completed the style check for frrbot

Thanks. No need to split the style fixes into a new commit. It is better to just squash that into the previous commit and pretend that you never had style issues ;).

Okay, thank you very much for your patient advice. I've now squashed it into a single commit. Please let me know if this looks good.

[Jafaral]

Unrelated CI failures, just triggered a re run of the failed tests. also, we should ignore the verify source failure. The reported style issues is consistent with other instances in the file. I.e, inline declaration of that variable doesn't have a new line in other places.