Fix/deadlock due to TCS setresult behaviour

[asdacap]

Fixes #5043

• It turns out TaskCompletionSource.SetResult may run it's continuation (task that is waiting for its result) synchronously. See this.
• So this means, when MSMS activate a forward sync feed via SyncDispatcher.UpdateState which locks _feedStateManipulation, it's TrySetResult will continue SyncDispatcher.Start from line 162 where it wait for dormantTaskLocal synchronously.
• This execution which held _feedStateManipulation lock may continue further into MergeBlockDownloader.Dispatch and remains until it attempt to download something (anything async).
• If however, SyncPeerAllocation.Replaced is triggered before any download is queue (likely during ChainLevelHelper.GetNextHeaders as it is IO bound), the replaced event execution will get blocked at SyncDispatcher.UpdateState as _feedStateManipulation is not released.
• If the sync execution from TCS will continue without releasing the lock until sync dispatcher attempt to free the sync peer allocation, it will hang but it can't free because _allocationLock is held by the the peer upgrade event.
• Running DoDispatch with Task.Run or setting TaskCreationOptions.RunContinuationsAsynchronously will work.

Changes:

• Run DoDispatch with Task.Run.
• Add TaskCreationOptions.RunContinuationsAsynchronously to both SyncDispatcher._dormantStateTask and Request.CompletionSource for good measure.

Types of changes

What types of changes does your code introduce?
Put an x in the boxes that apply

• Bugfix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• Documentation Update
• Code style update (formatting, renaming)
• Refactoring (no functional changes, no api changes)
• Build related changes
• Other (please describe):

Testing

Requires testing

• Yes
• No

In case you checked yes, did you write tests??

• Yes
• No

Comments about testing , should you have some (optional)

• goerli can sync.
• no clear indication of regression or improvement.

Further comments (optional)

If this is a relatively large or complex change, kick off the discussion by explaining why you chose the solution you did and what alternatives you considered, etc...