docs: removing nemotron and tildeopen model references

feat: lowered inference steps

feat: A0 removed and default to A2

test: remove stale test_text_engine.py (tests deprecated TextEngine class)

refactor: extract TTSEngine and ImageGenEngine from engine.py

feat: extract ImageGenEngine into core/image_gen.py

feat: extract TTSEngine into core/audio_gen.py

feat: cards work

fix: more adjustments for images

fix: image prompt and size

feat: reduce image generation size to 450x300

feat: accept instruct parameter in TTSEngine.synthesize()

fix: prompt fixes with differing types

feat: download correct model and creates the audio

feat: wire TTS into Phase 2 pipeline with voice design mode

feat: add voice design mode to TTSEngine with female speaking instruction

fix: target language and update docs

feat: update to use local langaguage and added drop down for more languages

fix: prompt adjustments

fix: dynamic cefr levels

fix: translation does work

feat: implement pipeline.py Phase 2 translation orchestration

feat: add retry loop to LlamaCppTextEngine for translation validation

docs: up to date docs with phase one done

feat: phase 1 is good enough

feat: phase 1 works

refactor: delegate MiniCPMTextEngine.generate to core.text_gen.generate_sentences

feat: add extract_sentences and generate_sentences to core.text_gen

feat: adjusting for type safety

feat: wrap MiniCPMTextEngine.generate() in retry loop for exact sentence count enforcement

feat: add _validate_lines and _build_retry_prompt helpers to MiniCPMTextEngine

fix: using cuda for llama cpp python

feat: replace TextEngine with MiniCPMTextEngine (llama-cpp-python, chat template)

refactor: rename nemotron_model_path to minicpm_model_path, remove MoE/ubatch params

feat: better feedback for errors

feat: cli params for moe

test: add inline tests for TextEngine generation and translation prompts

fix: comment out tildeopen

feat: swap tildeopen for tiny aya water for translation

feat: add EnginePool singleton — manages mutual exclusion between GPU engines

feat: add ImageGenEngine — diffusers Flux2KleinPipeline wrapper with lazy load/unload

feat: add TTSEngine — OmniVoice Python package wrapper with lazy load/unload

feat: add TextEngine — llama-cli subprocess wrapper for Nemotron and TildeOpen

feat: add Pydantic types for CardData, TextResult, AudioResult, ImageResult, EngineConfig

restructure: move app.py to root, replace model submodules with HF Hub download script, add package markers

feat: scaffold repo structure with empty files and headers