The difference between al-Sabi'un in Quran 5:69 and al-Sabi'in in Quran 2:62 is deliberate and grammatically defensible. It does not establish a grammatical mistake, although it does require recognizing that the two verses use different syntactic constructions.
Why the objection seems persuasive
In 5:69, the verse begins with inna: "Indeed, those who believe, those who are Jews, the Sabians, and the Christians..." Since inna normally puts its noun in the accusative, a critic expects al-Sabi'in, as it appears in 2:62. On a simple reading, al-Sabi'un appears to break the rule.
Indeed, those who believe, those who are Jews, the Sabians, and the Christians - whoever believes in Allah and the Last Day and does righteousness, they shall have their reward with their Lord. No fear shall be upon them, nor shall they grieve.
Quran 5:69
What the verse actually permits
The alleged error depends on assuming that al-Sabi'un must be coordinated directly with the noun governed by inna. Classical Arabic grammar does not require that analysis. The word may be read as a new subject, with its predicate understood from the clause that follows. The sense is: "Indeed, those who believe, those who are Jews, and the Christians [whoever believes in Allah...] - and the Sabians likewise [whoever believes in Allah...]."
Another accepted grammatical analysis treats al-Sabi'un as nominative because it follows the syntactic position, rather than the visible case ending, of the preceding construction. Classical grammarians and exegetes discussed both types of analysis. The point is not that the rules have been suspended, but that the sentence has been parsed differently from the simpler construction in 2:62.
Indeed, those who believe, those who are Jews, the Christians, and the Sabians - whoever believes in Allah and the Last Day and does righteousness, they shall have their reward with their Lord. No fear shall be upon them, nor shall they grieve.
Quran 2:62
Why the two forms are not contradictory
In 2:62, al-Sabi'in is straightforwardly included among the nouns governed by inna. In 5:69, the change in case signals a different arrangement, separating the Sabians from the immediately preceding sequence and then applying the qualifying statement to them as well. Arabic, like other highly inflected languages, can use case variation to indicate structure, emphasis, and rhetorical pacing.
A reader may prefer one classical parsing over another, and the unusual construction may reasonably prompt a grammatical question. But it is not a plain error. At most, the objection shows that 5:69 is rhetorically and syntactically more marked than 2:62. It does not show that the Quran violates Arabic grammar or that its meaning is defective.