Improve chained async exception/cancellation propagation

[benaadams]

Throwing and catching an exception isn't the cheapest of operations; which has been mentioned as a separate issue #12892 (comment)

However the way async propagates a thrown exception (or cancellation, which is via exceptions), is a chain of try/catch/throws; so the cost of an exception is multiplied by the number of awaits in the chain.

Could the propagation of exceptions in the common pattern of a chain of await -> await -> await be more efficient (avoiding the rethrow and catch step) to reduce this multiplication factor?

I'm mostly thinking for the simpler case where they flow directly(ish) into the next Task and then also fault or cancel that.

e.g. have the exception passed through the awaiter/asyncmethodbuilder mechanism and stamped with the next bit of stack rather than invoking the full exception handling machinery

[Dotnet-GitSync-Bot]

I couldn't figure out the best area label to add to this issue. If you have write-permissions please help me learn by adding exactly one area label.

[timcassell]

Related: dotnet/csharplang#4565, and dotnet/roslyn#65863

[mangod9]

related to some experimentation for 8.