fix(tui_gateway): stop live profile session reuse from hiding verification answers

[fangliquanflq]

What does this PR do?

Warm/live reuse of a non-launch profile session was reading the process launch state.db when building the user-visible transcript, so verification-candidate rows that only exist in the profile DB disappeared even though a cold resume of the same session still showed them. Route _live_session_payload through the existing profile-aware _session_db(session) helper so all surfaces agree.

Bug Cause

_live_session_payload hard-coded _get_db() (launch SessionDB). App-global remote profile sessions persist display history (including finish_reason=verification_required candidates) under session["profile_home"]/state.db. A miss on the launch DB fell back to collapsed in-memory model history, which already dropped those candidates.

Reproduction Steps

1. In desktop app-global remote mode, resume a session owned by a non-launch profile that has a persisted verification candidate.
2. Confirm cold session.resume shows the candidate.
3. Hit the live fast path via session.activate or a second session.resume that reuses the in-memory session.

Expected: candidate remains in the payload, matching cold resume.
Before fix: candidate missing; only the collapsed model history remains.

Fix

Use with _session_db(session) as db: before _live_visible_history so profile sessions open the correct DB and launch sessions keep the shared handle. Added test_live_session_payload_reads_profile_db_not_launch_db.

Related Issue

No issue

Type of Change

• Bug fix (non-breaking change that fixes an issue)

Changes Made

• tui_gateway/server.py - _live_session_payload reads history via _session_db(session) instead of launch _get_db()
• tests/test_tui_gateway_server.py - regression: profile DB has candidate, launch DB does not, live payload keeps candidate

How to Test

1. Manual: follow reproduction; after fix, live reuse matches cold resume.
2. Automated (already run locally):

scripts/run_tests.sh tests/test_tui_gateway_server.py -k "live_visible_history or live_session_payload or reconcile_display or session_resume_live" -q
scripts/run_tests.sh tests/tui_gateway/test_failed_turn_retention.py -q

Checklist

Code

• I've read the Contributing Guide
• My commit messages follow Conventional Commits
• I searched for existing PRs to make sure this isn't a duplicate
• My PR contains only changes related to this fix/feature
• I've run scripts/run_tests.sh on relevant tests and they pass
• I've added tests for my changes
• I've tested on my platform: Windows 11

Documentation & Housekeeping

• I've updated relevant documentation - N/A
• I've updated cli-config.yaml.example if I added/changed config keys - N/A
• I've updated CONTRIBUTING.md or AGENTS.md if I changed architecture or workflows - N/A
• I've considered cross-platform impact - N/A (DB path logic is cross-platform)
• I've updated tool descriptions/schemas if I changed tool behavior - N/A