I tend to believe most don’t understand the distinction between gain, level adjustment, and volume control. I have a pair of MC 901 McIntosh mono amplifiers and to read what it does is very confusing for my tiny brain.
I’ve always believe a true gain control increases voltage into the circuits to appoint where you can audibly here unpleasant distorted sound. The level of intensity increases and finding the point just before this happens leads to the best quality in sound.
It’s an interesting topic; wish I knew more about it. I’m certainly no electrical engineer.
From what I (try to) understand, many devices have limited ranges of optimal operation (loading and voltages) for lowest distortion - which is not conducive to coercing it into different gain levels. This is certainly true for tubes.
I guess you could look into op-amps (gain bandwidth product) or PS Audio’s so-called "gain cells" if you wanted to explore variable-gain circuitry.
In phono stages with active circuitry (not just a SUT with multiple taps), one "trick" is to have optional extra gain stages switched in & out to facilitate multiple gain levels. That’s not something you’d want to do in a tube amp.
As usual in engineering - the choice for greater flexibility usually comes with a cost: greater complexity, expense, and possibly sub-optimal sound quality parameters (noise floor, distortion, bandwidth).
IMO for a tube amp, a few discrete input attenuator positions (bypassable) is a good solution, and a better alternative than variable gain - it very cleanly addresses the *most common* problem, which is hearing the tube rush noise floor of a tube preamp when your system has too much net downstream gain. Each position could use very high quality resistors (Z-foils?) and you'd only need like 3 positions for good effect.