This is pretty much a boolean now...let's make it one. Also note PhysicalConnection is passes as `this`, it can't be null.

This helps for real-time output on long-running or hanging tests.

make sure there is only **one** path that can cause endpoints to be added, and ensure that the activation happens after the constructor - removes race condition that can lead to abandoned endpoints; note as part of this, switch the server-snapshot to span-based

This fixes net462, but not netcoreapp test hangs

Marc's working on connection code so we'll plug these in after.

This simplifies internal logistics and our profiling a bit

Fix #838 - provide mechanism for app-code to compute hash-slot using the muxer's state; also avoid hash-slot overhead when not cluster

fix #880 - missing zrevrangebylex; note that to avoid confusion this automatically adjusts to inverted parameter min/max usage, just like zrangebyscore already does

key bugs resolved:

- make sure that we only push things into _writtenAwaitingResponse until we have actually written them; they are allowed to throw certain errors *before* write, which breaks the queue completely
- ensire that MOVED redirects are sents as "internal call"; this means that if this is a new server, it can be queued rather than dropped
- allow MOVED code to select a server that isn't "up" yet

additional tweaks:

- change accessibility on WriteImpl to ensure everything goes via WriteTo
- renamed some methods to clarify what the lock status is: ...InsideWriteLock, ...TakingWriteLock, etc
- fix bug where ExportConfiguration can fail if a result is nil (although this might be a false outcome)
- why did StringGetWithExpiryMessage specify NoRedirect?
- clarify messages for ProtocolFailure and MOVED

We've de-duped keys most places, and others are just calling with 0. Still some left, but this takes care of most

Also parallelizes most Config test for quicker runs - moved the unfriendly ones to Failover.