combat "mini-game" is in most campaigns well over 50% of the things people do and in some it's well over 90% so we cannot just brush that aside.
I think you're on the right track here, but I think you've misdiagnosed the problem.
The problem boils down to "the rogue was built to shine in the 'exploration' pillar of the game" but because over time many tables (and to be fair, adventure modules) have focused exclusively on combat, rogue has a design problem and needs to be "fixed."
I think the root of the problem is that some classes (mostly martials but also a lot of spellcasters) are built for shining in the "combat" pillar of the game, and "when you have a hammer, everything looks like a nail." The extra care that has been spent on combat options for these characters, as well as the extreme streamlining and simplification of the skills system, means the game has pushing people in the direction of the combat pillar at the expense of everything else.
Page 47 of the BECMI DMG suggested that when stocking a dungeon, 2 of 6 rooms should be empty, 2 of 6 should have monsters, 1 of 6 should have traps, and 1 of 6 should have a special feature. I would argue that "empty rooms" are part of exploration (they certainly can't be "combat") and I would also posit that "traps" and "special feature" are also meant to fall in the "exploration" pillar, that means it was explicitly stated that exploration is supposed to happen
twice as often as combat in BECMI.
Similarly if you look at the 1E DMG random dungeon stocking tables (on p.171) you find 60% of rooms should be empty, 10% a monster with no treasure, 15% monster and treasure, 5% special feature, 5% trip/track, 5% treasure. By my count that's 25% combat (monsters) and 75% exploration... or in other words, exploration is supposed to happen
three times as often as combat in 1E.
Compare to page 296 of the 5E DMG where a full 50% of dungeon chambers should have "Monsters" - 13% hazards, 5% obstacles, 12% traps, 4% tricks, 15% empty rooms. Now, as you accurately point out,
the explicit expectation in 5E is that at least 50% of the game should be "combat."
So I would suggest the problem is NOT that the rogue needs to be re-designed for combat in 5.5E/6E/whatever. The problem is instead that the rules need to come with a
drastic de-emphasis of combat (and that means a lot of care needs to be put into examples of how to do that; the 5E DMG is 320 pages but has just 4 pages on "Traps" and 2 pages on "Exploration" including a half-page illustration... compared to 5 pages on "Combat", 3 pages of "Combat Options" and 10 pages on "Creating a Monster" - that's 18 pages on combat alone versus 6 to traps/exploration).
I don't expect a "Wurst of Grimtooth's Traps"-sized entry on traps and dungeon dressing in the DMG to compete with the Monster Manual (let's be honest, monsters are more fun for players to look than traps), but combat has clearly gotten far more treatment in 5E and is explicitly supposed to happen more often than it did in previous editions, but that means the class that isn't combat-focused isn't getting the proper amount of support.
Again, the solution
isn't to give the rogue more combat support. The solution is to give the DM more support on "all things non-combat" and make it crystal clear that combat should be happening MUCH LESS than exploration. (Combat is dynamic. Combat is where a lot of classes are built to shine. Combat also makes 1 minute of "game time" take an hour or more of real time; exploration usually makes hours or days of game time happen in seconds of real time at most tables; if your game is 90% combat it means you're literally spending 99% of the time at the table handling less than 1% of the actual time elapsed in game).
In other words, the rules need to make sure to give the rogue an environment where they can shine as they were designed to do!
Yes, D&D comes from wargaming roots so I expect combat to be the most detailed part. But we've put so much work into simulating war, we're losing the rest of it.
