Remove BalanceFlag and use two named arguments in dfpitr_reader()

[glourencoffee]

Description

The class BalanceFlag is used as an argument to the function dfpitr_reader() to define which kind of statements should be read:

cvm.dfpitr_reader('file.zip', cvm.BalanceFlag.CONSOLIDATED) # 1
cvm.dfpitr_reader('file.zip', cvm.BalanceFlag.INDIVIDUAL) # 2
cvm.dfpitr_reader('file.zip', cvm.BalanceFlag.CONSOLIDATED|cvm.BalanceFlag.INDIVIDUAL) # 3, default

BalanceFlag is a redundant class, though, since there already exists a class BalanceType. However, using BalanceType in its place would require dfpitr_reader() to take an iterable of BalanceType, which is kinda ugly:

cvm.dfpitr_reader('file.zip', cvm.BalanceType.CONSOLIDATED) # 1
cvm.dfpitr_reader('file.zip', cvm.BalanceType.INDIVIDUAL) # 2
cvm.dfpitr_reader('file.zip', [cvm.BalanceType.CONSOLIDATED, cvm.BalanceFlag.INDIVIDUAL]) # 3, default

It's ugly because there are only two types of balance, individual and consolidated, so taking an iterable makes it seem as though a lot of items are to be passed in to the function, when it will be at most two.

A better approach is to change dfpitr_reader() such that it takes two boolean arguments, individual and consolidated, replacing BalanceFlag:

cvm.dfpitr_reader('file.zip', consolidated=True,  individual=False) # 1
cvm.dfpitr_reader('file.zip', consolidated=False, individual=True)  # 2
cvm.dfpitr_reader('file.zip', consolidated=True,  individual=True)  # 3, default
cvm.dfpitr_reader('file.zip', consolidated=False, individual=False) # 4

Along with the existing functionality of dfpitr_reader(), this refactor will thus allow four possibilities:

1. Reading only consolidated statements (if present)
2. Reading only individual statements (if present)
3. Reading both consolidated and individual statements (whichever are present)
4. Reading neither consolidated nor individual statements, in which case only the general information about the document is read