Expose ExprEval::size to Python

[Thomasb81]

Following PR is a tentative to answer to #998.

Method ExprEval::size has been added. But no test demonstrating it works :

The method has several argument with usage not obvious to guess

  uint64_t size(
      const any* object, bool& invalidValue, const any* inst, const any* pexpr,
      bool full /* false: use only last range size, true: use all ranges */,
      bool muteError = false);

[Thomasb81]

test script used at my end on

typedef struct { bit [7:0] opcode; bit [23:0] addr; } IR;

module test();

  IR a_var;

endmodule

import uhdm

vpiObj_typedef_list = []


s = uhdm.Serializer()
data = s.Restore('slpp_all/surelog.uhdm')

typedef_iterator = uhdm.vpi_iterate(uhdm.vpiTypedef,data[0])
while(True):
    vpiObj_typedef = uhdm.vpi_scan(typedef_iterator)
    if vpiObj_typedef is None:
        break
    vpiObj_typedef_list.append(vpiObj_typedef)
    print(uhdm.vpi_get_str(uhdm.vpiName,vpiObj_typedef))

exprEval = uhdm.ExprEval()
data,invalidValue = exprEval.size(
        vpiObj_typedef_list[-1],
        invalidValue=invalidValue,
        inst=None,
        pexpr=None,
        full=False,
        muteError=False
)
print(data,invalidValue)

It print
32 True

[alaindargelas]

@Thomasb81, Awesome!
LGTM, please convert to full PR.