[Type][Backend] Support fp16

[chhzh123]

Description

This PR adds fp16 support for both LLVM and HLS backend. The LLVM backend needs to first represent the float16 value in int16, and the HLS backend uses half to represent fp16.

Examples

def test_fp16_array():
    def kernel(A: float16[10]) -> float16[10]:
        B: float16[10]
        for i in range(10):
            B[i] = A[i] + 1
        return B

    s = allo.customize(kernel)
    assert "f16" in str(s.module)
    mod = s.build()
    A = np.random.rand(10).astype(np.float16)
    B = mod(A)
    np.testing.assert_allclose(B, A + 1, rtol=1e-5)

Checklist

• PR's title starts with a category (e.g. [Bugfix], [IR], [Builder], etc)
• Changes are complete (i.e. I finished coding on this PR)
• All changes have test coverage (It would be better to provide ~2 different test cases to test the robustness of your code)
• Code is well-documented