aina yoredes l'kach

The Rabbonon on nidda 50b respond to R' Yochanan ben Nuri that there's a big difference between neveilas of tahor (tuma chamura) and tumas ochlin (tuma kala) that "Lo im amarta b'tuma chamura sheken aino yoredes l'kach." When the gemara first quotes this line of the Rabonon, Rashi explains that to mean that there is no tuma chamura which requires machshava whereas there is tuma kalla that requires machshava. Then he says that the gemara will explain. The problem is that the gemara never explains that. The gemara on 51a attempts to give one explanation that it means "aino oseh k'yotze bo." The gemara doesn't like that so it answers that it's "aina tzricha hechsher." The gemara questions if that is indeed true and concludes "shum tuma chamura baolam." Rashi explains this to mean that we're sticking with our second answer but we're just saying that never does a tuma chamura need hechsher whereas often ochlin does require hechsher. If that's pshat in the gemara then why did Rashi first explain it as referring to machshava? The gemara never even entertained the possibility.

Tosafos d"h shum on 51a asks why the gemara didn't answer that. The question is a good question on its own but with Rashi it's really a pele. The Aruch Laner points out that you could have understood the gemara's final answer of "shum tuma chamura baolam" as talking about machshava and not hechsher so it's a third distinct answer and not just qualifying the second answer. That would answer Tosafos' question but the problem is that Rashi (d"h shum) clearly doesn't learn like that. He says that we're talking about hechsher and not machshava.

Tosafos d"h shum in nidda 51a asks