JIT: Set edge likelihood in SetTargetEdge, SetKindAndTargetEdge

[amanasifkhalid]

Part of #93020. When setting the successor edge of a block with one successor, the edge likelihood should always be 1.0. We can enforce this in BasicBlock methods that set bbTargetEdge, and simplify their call sites.

[ghost]

Tagging subscribers to this area: @JulieLeeMSFT, @jakobbotsch
See info in area-owners.md if you want to be subscribed.

Issue Details

---

Part of #93020. When setting the successor edge of a block with one successor, the edge likelihood should always be 1.0. We can enforce this in BasicBlock methods that set bbTargetEdge, and simplify their call sites.

Author:	amanasifkhalid
Assignees:	-
Labels:	area-CodeGen-coreclr
Milestone:	-

[amanasifkhalid]

cc @dotnet/jit-contrib, @AndyAyersMS PTAL. No diffs, though there is a large MinOpts TP regression in coreclr_tests on Linux x64. I imagine this is from us setting unconditional edges' likelihoods in a phase we previously didn't, and some subset of tests that stress this phase/pattern are now regressed, TP-wise. If you'd like, I can dig into this locally and identify the culprits.

[AndyAyersMS]

Since the TP regression is just on linux hosts it's probably some compiler weirdness... hopefully it goes away with PGO.