Fix PE loader to recursivly import symbols and call DllMain

[t-moe]

This PR fixes some of the issues with the PE loader:

• DLL Imports were not recursively handled (as also fixed by Relocate import symbol addresses for loaded PE files #886)
• DLLs were not properly relocated, when their preferred base address was already taken (as also fixed by Fix bug: DLL address still use ImageBase address in spite of changing lib_base. #801)
• Entry point of DLLs (DllMain) were not called
• API sets were not handled properly

Calling DllMain is imperative, in order to initialize constructors (e.g. c++ static initializers and other library init stuff).

The idea proposed by #801 with delaying loading the DLLs until the first symbol is called, raises the question of when to call DllMain(). In this PR I load the DLLs recursively at startup, and also call DLLmain, but I skip the virtual DLLs in the hope that it is faster than #801. Also, by skip loading the virtual DLLs I hope that we can avoid copying the api-*.dlls entirely (which caused build problems in #886).

Unfortunately I was not able to run the testcases on my machine. Can you run the workflow?

Credits go to @crass (#886). I've copied his idea with the init_import method.

Checklist

Which kind of PR do you create?

• This PR only contains minor fixes.
• This PR contains major feature update.
• This PR introduces a new function/api for Qiling Framework.

Coding convention?

• The new code conforms to Qiling Framework naming convention.
• The imports are arranged properly.
• Essential comments are added.
• The reference of the new code is pointed out.

Extra tests?

• No extra tests are needed for this PR.
• I have added enough tests for this PR.
• Tests will be added after some discussion and review.

Changelog?

• This PR doesn't need to update Changelog.
• Changelog will be updated after some proper review.
• Changelog has been updated in my PR.

Target branch?

• The target branch is dev branch.

One last thing

• I have read the contribution guide

---

[Enteee]

@wtdcode this was closed in error, same issue as #918 . Removing dev branches is probably not very good practice...

[wtdcode]

@wtdcode this was closed in error, same issue as #918 . Removing dev branches is probably not very good practice...

Sorry I really don't know removing a branch would close issues related. Wait @t-moe to create a new one.

[Enteee]

No worries. Just wanted to point it out. There might be more PRs that were closed this way.

[elicn]

@wtdcode, about 10 PRs were closed this way.
I thought it was intentional..

[wtdcode]

@wtdcode, about 10 PRs were closed this way. I thought it was intentional..

I would have a look, really sorry.

[elicn]

Changes were included as part of #1118 .
Thanks @t-moe !