Fix flaky test

[peterzhu2118]

The test_evaluating_a_variable_entirely_within_c test is flaky. If we do the following:

diff --git a/test/unit/context_test.rb b/test/unit/context_test.rb
index 64b1852..eb5cd93 100644
--- a/test/unit/context_test.rb
+++ b/test/unit/context_test.rb
@@ -44,6 +44,7 @@ class ContextTest < Minitest::Test

       c_call_trace = TracePoint.trace(:c_call) do |t|
         unless t.self == TracePoint || t.self.is_a?(TracePoint)
+          puts "#{t.defined_class} #{t.method_id}"
           called_c_method_count += 1
         end
       end

We can see output such as:

 current
Thread abort_on_exception=
Thread::Queue pop
 current
Thread abort_on_exception=
Thread::Queue pop
Liquid::Context c_evaluate

And the failure is:

  1) Failure:
ContextTest#test_evaluating_a_variable_entirely_within_c [/Users/peter/src/github.com/Shopify/liquid-c/test/unit/context_test.rb:64]:
Expected: 1
  Actual: 7

I was able to reproduce this on Ruby 2.5.8 using seed 7822 (the repro is flaky and requires multiple tries).

As you can see, there are various other methods called that are also C functions.

The solution is to ignore all C function calls unless the class it is called on starts with Liquid.

[peterzhu2118]

Looks like this flaky test is causing the build on master to fail: https://github.com/Shopify/liquid-c/runs/1342582317

(This failure was not introduced in #86, as it can be repro'd on 1dc9459 as well).