#6632: fix(cli-runner): replace {model} placeholder in resumeArgs

## Summary - Fixes `{model}` placeholder not being replaced in `resumeArgs` for CLI backends - Ensures configured model is used on resume calls instead of defaulting to sonnet ## Problem When using CLI backends with custom `resumeArgs` that include a `{model}` placeholder: ```json "resumeArgs": ["-p", "--resume", "{sessionId}", "--model", "{model}"] ``` The `{model}` placeholder was not being replaced, causing: 1. First message: `--model opus` ✅ (new session with correct model) 2. Subsequent messages: `--model {model}` ❌ (literal string, API error or default model) This made it impossible to maintain a specific model across resume calls. ## Solution Add `.replaceAll("{model}", normalizedModel)` to the resumeArgs processing in `cli-runner.ts`, matching how `{sessionId}` is already handled. ## Testing - Configured `claude-cli/opus` as primary model for an agent - First message created new session with Opus - Subsequent messages (resume) now correctly pass `--model opus` - Model usage in responses confirms Opus is used throughout 🤖 Generated with [Claude Code](https://claude.com/claude-code) <!-- greptile_comment --> <h2>Greptile Overview</h2> <h3>Greptile Summary</h3> This PR fixes a placeholder substitution bug in the CLI runner’s resume path. When `backend.resumeArgs` contained `{model}`, it was previously passed through as a literal string on resume executions; the change mirrors existing `{sessionId}` handling by also replacing `{model}` with the normalized model ID before building the final CLI argv. This keeps resume calls consistent with initial session creation, preventing unexpected fallback/default model selection or CLI errors when resuming. <h3>Confidence Score: 5/5</h3> - This PR is safe to merge with minimal risk. - Change is narrowly scoped to resume-argument placeholder substitution in `runCliAgent`, uses an already-computed `normalizedModel`, and mirrors existing `{sessionId}` replacement logic without affecting non-resume execution paths. - No files require special attention <!-- greptile_other_comments_section --> _{(3/5) Reply to the agent's comments like "Can you suggest a fix for this @greptileai?" or ask follow-up questions!} **Context used:** - Context from `dashboard` - CLAUDE.md ([source](https://app.greptile.com/review/custom-context?memory=fd949e91-5c3a-4ab5-90a1-cbe184fd6ce8)) - Context from `dashboard` - AGENTS.md ([source](https://app.greptile.com/review/custom-context?memory=0d0c8278-ef8e-4d6c-ab21-f5527e322f13)) <!-- /greptile_comment -->