fix: dynamic orchestrator status, fleet agent counts, and E2E test suite

[toadkicker]

Summary

• Fix dynamic orchestrator status: Replaced hardcoded Active status with live get_orchestrator_status Tauri command response in fleet view; status now reflects real backend state
• Fix fleet active agent count: Active badge now counts only compiled-source agents (always-available built-ins) rather than all agents; total badge shows the full count across all sources
• Fix pap:// address bar suggestions: Added live: true to list_agents mock agents so build_catalog() indexes them — agents without live=true are silently filtered by serde default
• Fix typed-block render flakiness: Increased submitAndAwaitTypedBlock timeout (8s→15s), capped CI Playwright workers at 4 to reduce WASM CPU contention, and added retries: 1 for residual timing variance
• Fix E2E mock alignment: Updated canvas_plan_prompt handler (renamed from canvas_prompt), added 2 compiled-source agents, updated all fleet count / source badge / selector assertions to match current UI

Test plan

• 155 passed, 32 skipped (live-mode), 0 failed on full E2E suite locally with CI=true 4-worker config
• All typed-block schema templates verified (Movie, Book, WeatherForecast, GeoCoordinates, MusicRecording, SportsTeam, etc.)
• pap:// autocomplete suggestions appear correctly when typing pap:// prefix
• Fleet page shows correct ACTIVE (compiled) and TOTAL counts
• Orchestrator status reflects backend response dynamically

🤖 Generated with Claude Code

[github-actions]

Benchmark Regression Report

PAP Protocol Benchmark Regression Check
========================================
Baseline: .bench-baseline/baseline.json
Threshold: 30%

  ed25519_keypair_generation                19.8 µs  (baseline: 19.7 µs, +0.4%)  [ok]
  did_key_derivation                         1.5 µs  (baseline: 1.5 µs, +1.8%)  [ok]
  mandate_create_sign                       23.9 µs  (baseline: 24.3 µs, -1.9%)  [ok]
  mandate_chain_verify_depth3              126.4 µs  (baseline: 125.4 µs, +0.7%)  [ok]
  sd_jwt_issue_5claims                      27.9 µs  (baseline: 27.9 µs, -0.1%)  [ok]
  sd_jwt_verify_disclose_3of5               44.1 µs  (baseline: 44.4 µs, -0.4%)  [ok]
  session_open_full_lifecycle              108.6 µs  (baseline: 108.5 µs, +0.0%)  [ok]
  receipt_create_cosign                     48.2 µs  (baseline: 47.7 µs, +0.9%)  [ok]
  federation_announce_local                 56.2 µs  (baseline: 59.1 µs, -4.8%)  [ok]

All benchmarks within 30% of baseline.

_{Threshold: 10% regression vs baseline from main}