My apologies. I meant in the broader fantasy gaming genre, which does often have classes.
Broader as in CRPGS &MMOs would be more Tank/DPS/Healer/
Crowd-Control than warrior/mage/rogue or warrior/mage/priest. And, while the rogue might be DPS, so might a warrior or mage.
But why do you think that the rogue, warrior, and mage archetypes have become so pervasive
Relative to the original trinity of magic-user, fighter, and cleric, because the cleric was so counter-genre.
if the rogue/warrior distinction was really as superficial and redundant as you make it out to be?
The distinction isn't redundant, it's the niche protection, the walling off of practical abilities, that was uncalled-for. The distinction between a knight in shining armor and a dashing rogue is certainly legitimate, but that distinction doesn't call for the dashing rogue to be a pushover in a fight nor the knight to be incompetent off the battlefield. But that's what implementing the distinction by adding the Thief class did, to both of them.
Do you think that the thief spontaneously generated out of thin air in 1975 without any preexisting influences?
Like the ranger (Aragorn) and monk (Kwai Chang Caine), the thief had one very clear precedent: Lieber's Grey Mouser. Thus the oddities like 'Read Languages,' and preference for the sling, that otherwise didn't fit the 'Thief' concept so well.
There are different archetypes that the rogue and fighter are attempting to emulate, which is often reflected in D&D and elsewhere in other fantasy classes.
D&D's influence in the limited realm of fantasy RPGs can't be understated, but it doesn't mean it's mistakes must also be emulated. D&D has successfully moved the rogue forward from it's deplorable, wussy, niche-protected, incompetence-creating Thief origins to a reasonably well-rounded class concept. The contrast with the plight of the fighter, and the perspective that initial division of the two classes deepened those issues is informative, I think. But it's not a basis for nixxing the rogue from 5e at this late date.
More broadly, there are many archetypes that the abilities and trappings of the rogue and fighter, in some combination, come close to emulating, but not so many that they do so, individually, especially the poor fighter.
It sounds like you are turning the rogue into your scapegoat for the situation of the fighter.
That would imply my intent was to excuse it. The fighter and rogue started with similar issues, the rogue has come further as the game evolved and retained more of it's progress in 5e than did the fighter.
an aside, it amazes me that you can advocate for the existence for the warlord out of the same mouth that you advocate against the rogue.
I'm very versatile.

They're not as contradictory as you might think, though, a number of classes - the rogue, monk, barbarian, paladin and ranger - could all arguably be superfluous if the fighter had gotten a fairer shake from the beginning, but they're still in the game, even where the fighter comes closest to covering their schtick. The 5e fighter doesn't come anywhere near covering for the Warlord. So, it should really be among those other arguably-superfluous classes in the PH.
Not that you couldn't fold virtually all the Warlord's toys in with the fighter's & rogue's and still not have a class to assail Tier 1, nor even Tier 2.
In a broader TTRPG context, ideally, if games were to use classes at all, a 'Hero' that subsumed every martial class D&D has ever presented might be most appropriate.