Fix issue in accept sequence

[shubhamugare]

The accept sequences with ignore terminals were resulting in retaining grammatically incorrect tokens in some cases.

Test:

python3 syncode/infer.py --mode grammar_mask --model meta-llama/Llama-2-7b-hf  --dataset humaneval --grammar python --max_new_tokens 400 --parser lr --parse_output_only False

results in pass@1 of ~13.41%

python3 syncode/infer.py --mode grammar_strict --model google/gemma-2-2b-it --dataset json_eval --grammar json --max_new_tokens 400 --parser lalr --device cuda:1

results in 98% accuracy

Fixes #143