[Feature] Support L1 calling convention in tvm-ffi

The current calling convention in Tilus:

import tilus
from tilus import float16, int32, GlobalTensor


class Example(tilus.Script):
    def __call__(
        self, m_size: int32, n_size: int, k_size: int, a_ptr: ~flaot16, b_ptr: ~flaot16, c_ptr: ~flaot16
    ):  # m_size is dynamic, n_size and k_size are JIT constants
        """
        Generate:
            void example(int m, int n, int k, float16 *a, float16 *b, float16 *c) {...}
        """
        ...
        g_a: GlobalTensor = self.global_view(a_ptr, dtype=float16, shape=[m_size, k_size])
        ...

The L1 calling convention:

class ExampleL1(tilus.Script):
    def __call__(
        self, 
        g_a: GlobalTensor[float16, ['m', 'K']],  # m is dynamic,
        g_b: GlobalTensor[float16, ['K', 'N']],  # K and N are JIT constants
        g_c: GlobalTensor[float16, ['m', 'N']],
    ):
        """
        Generate:
            void exampleL1(tvm::ffi::TensorView g_a, tvm::ffi::TensorView g_b, tvm::ffi::TensorView g_c) {...}
        """
        ...

We can also use

        g_a: 'float16[m, K]',
        g_b: 'float16[K, N]',
        g_c: 'float16[m, N]'

as type annotation.

No files or tests are named. Start by locating the existing Tilus calling-convention and GlobalTensor type-annotation entry points, then trace how kernel parameters become generated signatures. Done means supporting the L1 TensorView convention with both shown annotation forms while preserving the current convention, with tests covering dynamic and JIT-constant dimensions.