docs(ton-bridge): clarify auto-loading and optional config

[konard]

Summary

Fixes #27

Root cause analysis

The "not started" message users see is the LLM's response when it can't find the ton_bridge_* tools. The teleton-agent plugin-loader automatically loads all plugins found in ~/.teleton/plugins/ — no plugins: section entry is needed in config.yaml.

The plugin code itself is correct (all 25 tests pass, validation passes). The confusion arose because:

1. The README didn't clearly explain that plugins are auto-loaded
2. The README didn't clarify that the plugins: config section is optional (defaults work out-of-the-box)
3. The section ordering didn't match other plugins (usage examples were after config)

Changes

plugins/ton-bridge/README.md

• Added note after install command: "Restart Teleton — the plugin is auto-loaded from ~/.teleton/plugins/. No changes to config.yaml are required."
• Moved usage examples before configuration section (matches casino, ton-trading-bot pattern)
• Marked configuration as optional with explicit note: "Configuration is optional — the plugin works out of the box with defaults."
• Updated section naming to match other plugins (Install, Usage examples, Tool schemas)

How the loader works (for reference)

From teleton-agent/src/agent/tools/plugin-loader.ts:

const pluginConfigKey = pluginName.replace(/-/g, "_");
// "ton-bridge" → "ton_bridge"
const rawPluginConfig = (config.plugins?.[pluginConfigKey] as Record<string, unknown>) ?? {};
const pluginConfig = { ...manifest?.defaultConfig, ...rawPluginConfig };
// Works without any config.yaml entry — uses manifest.defaultConfig

The plugins: section in config.yaml is only needed to override defaults, not to enable the plugin.

Test plan

• All 25 existing tests pass (npm test)
• Plugin validation passes (npm run validate — [OK] ton-bridge: 3 tool(s) validated)
• No code changes — only README documentation update

🤖 Generated with Claude Code

[konard]

🤖 Solution Draft Log

This log file contains the complete execution trace of the AI solution draft process.

💰 Cost estimation:

• Public pricing estimate: $4.491921
• Calculated by Anthropic: $2.587566 USD
• Difference: $-1.904354 (-42.40%)

🤖 Models used:

• Tool: Claude
• Requested: sonnet
• Main model: Claude Sonnet 4.6 (ID: claude-sonnet-4-6, Anthropic, cutoff: 2025-08)
  📎 Log file uploaded as Gist (1638KB)
  🔗 View complete solution draft log

---

Now working session is ended, feel free to review and add any feedback on the solution draft.

[konard]

✅ Ready to merge

This pull request is now ready to be merged:

• All CI checks have passed
• No merge conflicts
• No pending changes

---

Monitored by hive-mind with --auto-restart-until-mergeable flag