build: Install tensorflow-macos on Apple silicon Macs (#2119)

* Use hatchling's environment markers system to use platform information
  to determine when the installing machine is an Apple silicon Mac
  and install tensorflow-macos instead of tensorflow in this situation.
   - c.f. https://hatch.pypa.io/latest/config/dependency/#environment-markers
* This works because of the following:
   - x86-64 Linux machine:
     >>> import platform
     >>> platform.machine()
     'x86_64'
     >>> platform.system()
     'Linux'
   - x86-64 Apple machine:
     >>> import platform
     >>> platform.machine()
     'x86_64'
     >>> platform.system()
     'Darwin'
   - AArch64 machine:
     >>> import platform
     >>> platform.machine()
     'aarch64'
     >>> platform.system()
     'Linux'
   - Apple silicon machine:
     >>> import platform
     >>> platform.machine()
     'arm64'
     >>> platform.system()
     'Darwin'

pyproject.toml

@@ -66,7 +66,8 @@ Homepage = "https://github.com/scikit-hep/pyhf"
 [project.optional-dependencies]
 shellcomplete = ["click_completion"]
 tensorflow = [
-    "tensorflow>=2.7.0",  # c.f. PR #1962
+    "tensorflow>=2.7.0; platform_machine != 'arm64'",  # c.f. PR #1962
+    "tensorflow-macos>=2.7.0; platform_machine == 'arm64' and platform_system == 'Darwin'",  # c.f. PR #2119
     "tensorflow-probability>=0.11.0",  # c.f. PR #1657
 ]
 torch = ["torch>=1.10.0"]  # c.f. PR #1657