feat: wire blast-radius to /v1/analysis/impact endpoint

[jonathanpopham]

Problem

The blast-radius command currently calls /v1/graphs/supermodel, gets a generic graph, and does naive reverse BFS on import edges in Go. This misses the API's real impact analysis which includes:

• Call graph + dependency graph reachability (not just imports)
• Risk scoring (low/medium/high/critical) with risk factors
• Affected functions with distance and relationship type (direct_caller/transitive_caller)
• Affected entry points (route handlers, module exports, event handlers)
• Diff-based analysis (auto-extract targets from git diff)
• Global coupling map when no targets specified
• Cross-domain impact detection

The API already exposes all of this at POST /v1/analysis/impact.

Solution

Wire the CLI's blast-radius command to the dedicated endpoint:

1. Add Impact() method to api.Client that POSTs to /v1/analysis/impact with async polling
2. Add response types (ImpactResult, BlastRadius, AffectedFunction, AffectedFile, etc.)
3. Rewrite internal/blastradius/handler.go to call the new endpoint
4. Support --targets query param and --diff file upload
5. Surface risk scores, affected functions, and entry points in output

Three modes

1. supermodel blast-radius <file> — target a specific file (maps to ?targets=<file>)
2. supermodel blast-radius --diff changes.diff — analyze from a diff
3. supermodel blast-radius (no args) — global coupling map

References

• API spec: POST /v1/analysis/impact in openapi.yaml
• Current CLI handler: internal/blastradius/handler.go (naive import BFS)
• Pattern: internal/deadcode/handler.go (already wired to dedicated endpoint)

[coderabbitai]

🔗 Related PRs

#15 - fix: use correct API endpoint /v1/graphs/supermodel [closed]
#18 - feat: add supermodel restore command [closed]
#20 - feat: supermodel factory — AI-native SDLC orchestration [closed]
#23 - feat: wire dead-code to /v1/analysis/dead-code endpoint [closed]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!