fix: prevent concurrent dials

[joshuef]

No description provided.

[reviewpad]

AI-Generated Summary: This pull request introduces a fix to prevent concurrent dials in the safenode network implementation. The changes involve modifying the SwarmDriver implementation to immediately update the pending_dial hashmap when launching a new dial, as dials can take time and may cause conflicts if not managed properly. Additionally, error handling has been improved by informing the user of any dial errors encountered during the process.

[reviewpad]

Reviewpad Report

⚠️ Warnings

• Please link an issue to the pull request

‼️ Errors

• Unconventional commit detected: 'Move log to make sense when adding peer to routing table' (1d992f3)

[reviewpad]

AI-Generated Summary: This pull request consists of two patches that aim to improve the network handling in a SafeNode application. The first patch fixes an issue of preventing concurrent dials by updating the pending dial hashmap immediately and handling dial errors properly. The second patch moves a log statement about adding a peer to the routing table to a more appropriate location in the event handling logic.