Append optnone and noinline for -O0 optlevel

[kvpanch]

This appears to be NFC change as new pass manager handles --passes=default correctly. Still must be useful as a safeguard and possibly for the old pass manager that is still used by backend.

Contract                        Without (bytes)   With noinline+optnone (bytes)
─────────────────────────────── ───────────────── ────────────────────────────
contract.sol:C                            1,266                          1,266
Fibonacci:Iterative                       2,506                          2,506
Storage                                   3,671                          3,671
Fibonacci:Recursive                       3,794                          3,794
Computation                               5,776                          5,776
Fibonacci:Binet                           8,786                          8,786
large_div_rem                            10,705                          10,705
DivisionArithmetics                      14,831                          14,831
ERC20                                    32,288                          32,288
ERC20Tester                              34,589                          34,589

[kvpanch]

still not clear why in @elle-j 's example there were SROA and mem2reg and why CSE optimized 2 identical loads, which is what I don't see for the simple c program https://godbolt.org/z/zajPn5e4M

[elle-j]

...why CSE optimized 2 identical loads, which is what I don't see for the simple c program

You can see the optimization of the loads here:
https://godbolt.org/z/ax4nhbzMf:

And to see the one DAG node (t5), see:
https://godbolt.org/z/bYba7jczr

Initial selection DAG: %bb.0 'square64:'
SelectionDAG has 10 nodes:
  t0: ch,glue = EntryToken
  t3: i64 = Constant<0>
    t2: i64,ch = CopyFromReg t0, Register:i64 %0
  t5: i64,ch = load<(load (s64) from %ir.p, align 32)> t0, t2, undef:i64
    t6: i64 = mul t5, t5

*** MachineFunction at end of ISel ***
# Machine code for function square64: IsSSA, TracksLiveness
Function Live Ins: $x10 in %0

bb.0 (%ir-block.0):
  liveins: $x10
  %0:gpr = COPY $x10
  %1:gpr = LD %0:gpr, 0 :: (load (s64) from %ir.p, align 32)
  %2:gpr = MUL %1:gpr, %1:gpr
  $x10 = COPY %2:gpr
  PseudoRET implicit $x10

mem2reg

I don't believe I saw this applied in one of the earlier examples, that was for some higher opt level.