DNM prototype for lower limit message buffer

[fjetter]

cc @wence- This is roughly what I had originally in mind (there are still a couple of problems to iron out but overall it works) regarding #7664

I'm mildly disappointed since the simple test I was trying this with didn't change in terms of performance but I haven't run our entire benchmark suite

[github-actions]

Unit Test Results

See test report for an extended history of previous test failures. This is useful for diagnosing flaky tests.

       25 files   -        1         25 suites   - 1   12h 59m 35s ⏱️ + 27m 34s
  3 545 tests ±       0    3 431 ✔️  -        2     105 💤 ±    0    9 ❌ +  2 
43 114 runs   - 1 711  41 162 ✔️  - 1 548  1 921 💤  - 186  31 ❌ +23 

For more details on these failures, see this check.

Results for commit 06ffb3a. ± Comparison against base commit 51ce7cf.

[wence-]

Thanks, I'll have a look through properly to compare.