Retry every client connection failure; make long outages cheap

[andrewchen5678]

Summary

• The CLI client no longer exits when its first connection attempt fails on a transient error. A failed attempt is retried exactly like a lost connection; only permanent errors (rejected key, bad config, proxy port in use) end the session. --no-auto-reconnect and --max-reconnect-attempts keep their names but now cover the first attempt too.
• Long outages are cheap: backoff cap raised from 60s to 5 min, and the endpoint rebuild still fires after the third consecutive failure but then repeats at most every 30 min per outage instead of on every third attempt.
• On iOS, the app coming to the foreground ends a backoff wait immediately with a fresh series, as a restored network path already did.
• The reconnect loop publishes failed attempts, last error, and next attempt time; the CLI status snapshot carries it and the control panel shows it on the phase line.
• Docs, config example, flag help, and FFI comments updated. Version bumped to 0.0.79.

Test plan

• ci/unix/ci.sh (clippy -D warnings + core and CLI tests) green; workspace clippy including desktop clean
• New unit tests: backoff cap, rebuild gating, first-failure retry, retry cap, foreground wake, parked wait ignores foreground
• New hermetic e2e test: client started against a server that rejects until a late-start signal survives and connects
• Live runs with the built binaries: client started before server waits through a 30s connect timeout and connects when the server appears; SIGKILLed server is detected and reconnected on restart with proxy traffic flowing both times; --no-auto-reconnect and --max-reconnect-attempts 1 exit nonzero as documented; unauthorized key exits immediately without retry
• ./build-ios.sh release rebuilt the xcframework on the Mac VM

🤖 Generated with Claude Code