Improved performance of rz_bv_copy_nbits and rz_bv_set_range

[rajRishi22]

Your checklist for this pull request

• I've read the guidelines for contributing to this repository
• I made sure to follow the project's coding style
• I've documented or updated the documentation of every function and struct this PR changes. If not so I've explained why.
• I've added tests that prove my fix is effective or that my feature works (if possible)
• I've updated the rizin book with the relevant information (if needed)

Detailed description
This pull request optimizes the performance of the rz_bv_set_range function. The original implementation iterated through bits one at a time, leading to inefficiencies for large ranges. The updated implementation:

Processes aligned chunks of bits using system word size for faster operations.
Dynamically adjusts chunk size for different architectures (e.g., 32-bit, 64-bit).
Handles unaligned prefix and suffix bits separately while optimizing the main loop.
Adds robust boundary validation to ensure correctness.
This change reduces iteration overhead and improves performance while maintaining compatibility and correctness.

Test plan

1. Verify functionality for small ranges, large ranges, and edge cases (unaligned ranges).
2. Test on various architectures to confirm portability (32-bit, 64-bit).
3. Use unit tests to ensure the results match the original functionality.
4. Check for any regressions with existing test suites.

Closing issues
Partially addresses #4716
...

[Rot127]

Nice start!
I can imagine we can even optimize the unaligned cases. But let's do this later.

I also changed your PR message. Because it closes #4716 only partially (missing unaligned cases).

You can add test cases in test/unit/test_bitvector.c.
Once everything is implemented and passes, we can run the Travis CI to test it on big endian machines.

[wargio]

missing implementation of rz_bv_get_chunk

[wargio]

Suggested change

	first_word &= ~(0xFFFFFFFF >> bit_offset); // Clear the upper bits
	second_word &= (0xFFFFFFFF >> (32 - bit_offset)); // Clear the lower bits
	first_word &= ~(UT32_MAX >> bit_offset); // Clear the upper bits
	second_word &= (UT32_MAX >> (32 - bit_offset)); // Clear the lower bits

[wargio]

You are touching something very critical within rizin, so please provide benchmark and C test for correctness and edge cases.