runtime/v2: net.Dial gRPC shim sockets before trying grpc

[dcantah]

This is mostly to workaround an issue with gRPC based shims after containerd restart. If a shim dies while containerd is also down/restarting, on reboot grpc.DialContext with our current set of DialOptions will make us wait for 100 seconds per shim even if the socket no longer exists or has no listener.

[fuweid]

LGTM

REF: #6031

[dcantah]

/cherrypick release/1.7

[k8s-infra-cherrypick-robot]

@dcantah: #9447 failed to apply on top of branch "release/1.7":

Applying: runtime/v2: net.Dial gRPC shim sockets before trying grpc
Using index info to reconstruct a base tree...
M	runtime/v2/shim.go
Falling back to patching base and 3-way merge...
Auto-merging runtime/v2/shim.go
CONFLICT (content): Merge conflict in runtime/v2/shim.go
error: Failed to merge in the changes.
hint: Use 'git am --show-current-patch=diff' to see the failed patch
Patch failed at 0001 runtime/v2: net.Dial gRPC shim sockets before trying grpc
When you have resolved this problem, run "git am --continue".
If you prefer to skip this patch, run "git am --skip" instead.
To restore the original branch and stop patching, run "git am --abort".

In response to this:

/cherrypick release/1.7

Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository.