Using hatchling produces better intellisense with vscode (#200)

Summary:
The intellisense has been messed up in vscode for fp8_experimental and conda envs, I found that changing to hatchling fixes this issues.

Pull Request resolved: #200

Reviewed By: bdhirsh

Differential Revision: D53457194

Pulled By: drisspg

fbshipit-source-id: a2f6d8f2dab1d2491f327a9bbabc439b5f17e8ed

Author: drisspg

pyproject.toml

@@ -1,8 +1,6 @@
-# pyproject.toml
-
 [build-system]
-requires = ["setuptools>=61.0"]
-build-backend = "setuptools.build_meta"
+requires = ["hatchling"]
+build-backend = "hatchling.build"
 
 [project]
 name = "float8_experimental"