fromIteratorBuffered

[Avasil]

Fixes #1275

I've run a simple benchmark against fromIterator + bufferSliding and there seems to be a decent gain (around 35%, perhaps more than 3.2.2 because we fixed perf regression on Scala 2.13) but I'm not sure if it is enough to make a difference for your use case @deusaquilus

If it's not, I'm happy to try and help to solve the issue

[info] Benchmark                                               (chunkCount)  (chunkSize)   Mode  Cnt    Score   Error  Units
[info] ObservableIteratorBenchmark.bufferTumbling                      1000         1000  thrpt   40   95.434 ± 0.998  ops/s
[info] ObservableIteratorBenchmark.fromIteratorBufferedUnsafe          1000         1000  thrpt   40  130.350 ± 2.840  ops/s

[deusaquilus]

@Avasil Actually it might be even more then this number in my use-case because the JDBC ResultSet fetches results in a chunked way itself. That means that by synchronizing Monix Chunks and JDBC chunks, we might be freeing a whole lot of IO cycles downstream. Will let you know how it turns out.

Btw what release can I hope to see this in?

[Avasil]

@deusaquilus It's already released in 3.3.0 :) Will be waiting for your report