With the help of the word generation and ambiguity resolution codes of the Zemberek library, I produced about 200000 derived Turkish words out of 10000 Turkish word roots. Then I threw them into csv.
*Thanks to https://github.com/ahmetaa/zemberek-nlp/blob/master/examples/src/main/java/zemberek/examples/morphology/GenerateWords.java and https://github.com/ozturkberkay/Zemberek-Python-Examples/blob/master/examples/morphology/word_generation.py
*Thanks to https://github.com/ahmetaa/zemberek-nlp/blob/master/examples/src/main/java/zemberek/examples/morphology/StemmingAndLemmatization.java and https://github.com/ozturkberkay/Zemberek-Python-Examples/tree/master/examples/morphology/stem-and-lemmatize