fix(core): implement robust auto-discovery for model layers

test(core): update HRM test case to correct structure

fix(core): correct layer paths for HRM and Qwen

chore: remove diagnostic logs

chore: update Dockerfile with new default model

chore: switch default model to Qwen/Qwen2.5-0.5B-Instruct

debug(core): add model structure logging

test(core): add HRM-style architecture test case

fix(core): add HRM-Text-1B layer path and attention attribute

test(generation): add unit tests for render_prompt

fix(generation): handle missing chat template and add HRM-Text-1B fallback

Switch default model to HRM-Text-1B

Fix sentence token alignment

Fix Docker build context

Deploy Thought Anchors

initial commit