Task 1: add Bio Player (pointwise-activation ESN) — 14.40 ± 0.36

[GuanchunLi]

Summary

Adds braincraft/env1_player_bio.py, a handcrafted pointwise-activation Echo State Network controller for Task 1, together with a full model description in braincraft/env1_player_bio.md.

No learning / no training loop — the weight matrices are fixed at build time and the single yield in bio_player() returns the frozen model.

Approach

Every hidden activation is a scalar function of its own preactivation; all cross-neuron logic lives in the connectivity matrices. Forty of the 1000 hidden slots are active; the rest are unused. The controller combines four behaviours:

• a reflex wall-follower (proximity / safety / hit taps),
• an initial-heading correction that cancels the ±5° start-direction
  perturbation so the integrated position (pos_x, pos_y) tracks true
  position,
• a rising-edge energy-reward detector that arms a shortcut once the
  bot has seen its first source,
• a pose-gated corridor shortcut that fires when the reward has been
  seen and the bot crosses a clear horizontal corridor, steering by
  sign(sin(phi)) · sign(pos_y).

See env1_player_bio.md for the full activation library, slot layout, wiring equations, and readout weights (eight non-zero Wout entries).

Assumed performance

Running python braincraft/env1_player_bio.py (seed 12345, 10 runs):