fix(tui): close slash_worker in _finalize_session to prevent zombie processes #38095

[kyssta-exe]

Summary

When a TUI session is closed, was not cleaning up the subprocess. This left orphaned processes accumulating each time the TUI was opened and closed while the dashboard remained running.

Problem

handles memory commits, notification stops, and DB session ending, but does not close the subprocess. Code paths that call without separately handling worker cleanup (e.g., TUI close without explicit RPC) leave zombie processes.

Fix

Add slash_worker cleanup to so any code path that calls it properly terminates the worker subprocess. The method is idempotent (checks before terminating), so double-close from is safe.

Testing

• Verified is idempotent
• Follows existing error-handling pattern (best-effort try/except)
• No lint errors introduced

Fixes #38095

[alt-glitch]

Likely duplicate of #38114 — both fix #38095 by closing session['slash_worker'] in _finalize_session() (tui_gateway/server.py). #38114 additionally hardens _SlashWorker.close() and nulls the ref in _restart_slash_worker, plus adds 5 tests (but bundles an unrelated version bump). Same zombie-leak family as #21370/#27073/#22863. Flagging for de-dup.

[teknium1]

Superseded by #42132 (merged), which closes the slash_worker subprocess leak via two guards: process-group kill on PTY teardown + a cross-platform parent-death watchdog in the worker. Closing as resolved — thanks for tackling this; the merged fix salvaged the process-group-kill and watchdog approaches with contributor authorship preserved.