Everything makes sense until mm. 5. It looks like it moves from c minor to Eb major.
Suppose we're in Eb major starting at mm. 5
From the end of mm. 5 to the very beginning of mm. 6, it's a vii°7/vi in Eb. It cannot be vii°7 because the root is B. It can't be CT°7 of the next chord because the root of the next chord is not in it.
Then it moves to the V7 chord. But it's not used functionally because it never resolves.
Next, it becomes even more interesting. It goes to vii°42/D and V42/D. And it resolves to D6 in mm. 7.
Finally, it goes to vii and V of c minor in mm. 8-9 (simplified a bit here).
mm. 5 is probably never in Eb
That D chord in mm. 7 is probably not in Eb major. It makes more sense to be a V/V in c minor.
If we work backward, that vii°7/vi in Eb should be vii°7 in c. It came from iv42 in c (instead of ii42 in Eb). Beethoven just used that long dominant chord in mm. 4 to tonicize to III of c minor.
But what is that unresolved dominant in mm. 6?
It probably makes sense in the context of a sequence. The voice leading takes precedence of its function. And it sounds surprising. I guess there can be a better explanation here. I'm curious to hear what people think.
Conclusion
Thanks @Aaron for the response. I think everything makes more sense taking into account the enharmonic spelling. So this is what I have. Everything is in c minor.
measure 5
It starts on III of c minor. Goes to iv42 on beat 3.
On beat 4, it moves from CT°7 -> iv64 -> vii°7/III.
measure 6
It moves from vii°7/III to V7/III on beat 2.
In the rest of the measure, instead of resolving the V7, it moves on from vii°42/V/V -> V42/V/V -> vii°43/V/V.
measure 7-8
It resolves that vii°43/V/V to V6/V.
Starting from beat 3 until the end of beat 2 on measure 8, it goes through inversions of vii°42 and lands on V7 in beat 3.
Finally, it moves from V7 -> iv7 -> V7.
At a high level
If we ignore those fancy applied chords (vii -> V on those tonicized centers), this is what happens from measure 5-8 (ignoring the inversions).
III -> iv -> V/III (or VII, unresolved) -> V/V -> V -> iv -> V -> i
I still find that unresolved dominant in measure 6 very interesting. It's Beethoven telling you "Hey, think I'll resolve to Eb? Guess where I go next!". In terms of listening, it creates a strong sense of searching.
Another interpretation
I originally tried to find a linear progression here but I couldn't because I was trying to find one that connects the entire passage.
If we only look at measure 5 beat 3 until measure 6 beat 3, it is a linear progression connecting iv42 and vii°42/V/V.
I also found https://www.harmony.org.uk/book/linear_progressions.htm, which has a Mozart example with a similarly unresolved V chord in the progression.
Some people believe that Beethoven was inspired by this Mozart sonata. I think that makes sense. There are many similarities between the two and Beethoven just takes it to the next level.
