model.ts: rename ONNX input ref_s → style for kokoro-js / kokoro-onnx interop

synthesize: strip BOS+EOS audio from output (keep BOS in model input)

synthesize: prepend BOS/EOS to inputIds (was missing — caused leading-consonant eat)

speed slider now scales pred_dur (was playback-rate only)

loanword preprocessor + re-exported v0.2 ONNX (torch 2.6)

release: webgpu demo + source