Rescue PR #14: one ego per scene (Daphne Apr 16 directive)

[smrifaki]

Context

Picking up Charlie's stale PR #14 one_ego_per_scene (open since 2025-11-24). Implements Daphne's Apr 16 directive: "train in a single-agent setting for now to keep the story clean (i.e., control one agent)."

Status

This PR is a direct push of Charlie's original branch (commits unchanged) so the work doesn't disappear when Charlie's account goes inactive. The branch has not been rebased onto current main.

The original PR #14 had 2 Greptile bot bugs flagged 2025-11-24 in drive.py lines 220 and 420 (AttributeError when co_player_conditioning is None). Both are FIXED in commits bc52eb7 and 52f3884 on this branch — already on top.

A third instance of the same pattern remains at drive.py:447 and should also be guarded:

# current
if self.co_player_condition_type != "none":
# proposed
if hasattr(self, 'co_player_condition_type') and self.co_player_condition_type != "none":

Merge strategy needed

Direct rebase onto current main is infeasible — current main has ~3.9M lines of PufferLib upstream sync changes since this branch diverged. The actual application-logic changes are small (~554 insertions, 164 deletions across 9 files), but they collide with PufferLib 3.0 sync changes in:

• pufferlib/vector.py
• pufferlib/ocean/drive/drive.py
• pufferlib/ocean/drive/binding.h
• pufferlib/config/ocean/adaptive.ini

Suggested next step: cherry-pick Charlie's commit 284c9d86 "Training with one ego per world working" onto current main with manual conflict resolution by someone who knows the PufferLib 3.0 changes (Mohit?).

Why open this draft

So the Daphne Apr 16 directive doesn't get lost. This branch preserves Charlie's work. Easier to port from than to find buried in 5-month-old commits.

Reference

• Original PR: One ego per scene #14 by @charliemolony 2025-11-24
• Daphne Apr 16 directive in #k-shot-agents Slack
• Bug catches by Greptile bot (2 of 3 already fixed in this branch)