As I was coming home and getting ready to check the thread again, I was just thinking it needed more blowhard "it can stall at any airspeed" posts, and when I refreshed it I was not disappointed.
In more serious matters, I saw Wiedhopf's plot of the granular FR24 speed and altitude data, and it gave me the idea to grab the spreadsheet myself and calculate the Total Energy and add it to the plot, to see if it might reveal anything. I converted the columns to meters per second (speed) and meters (altitude) for consistency of units and made a new cell that contains .5*speed^2+9.807*altitude, which is kinetic plus potential energy without the mass, which is approximately constant. (The result is divided by 5 anyway, to make the graph height roughly match the others. So it's an arbitrary unit, but we get to see the changes and trends.) There's also vertical speed from FR24, which I assume it auto generated from altitude over time. The black and blue colors for altitude and speed are preserved from Wiedhopf's graph, the VS is in green, and the TE in orange.
So what do we see in the TE? During the main steady-state climb, the TE is evenly increasing with a constant slope, just as we'd expect. And where there's the first big altitude dip and speed rise, there's barely a change in the TE slope, which is what we would expect for an even airspeed-altitude tradeoff (i.e., no significant thrust or drag changes, be it from a drag device or significant maneuvering.) A closer look does actually show a slight decrease, which matches a speed (i.e., drag) rise significantly above best L/D. So far so good, the plot is "sane."
Anyway, the small amount of downward kink in that slope, I think, allows us to calibrate our eye for further kinks later on. Once the main climb is done and they level off, all the altitude and VS meanders are smaller than during the original -2500 fpm dip, but the TE meanders are bigger. That shows to me, drag or thrust changes. I think we can rule out drag changes from drag devices as it would be unlikely any would be deployed for any reason, and we can also rule out drag changes from maneuvering, as the track plot shows more or less straight ahead, and the vertical meanders are all smaller than the first -2500 fpm one. By process of elimination, that leaves thrust changes, which is consistent with the present speculation of the crew trying to fly the airplane with degraded instrumentation or control difficulty.